Dissertations / Theses on the topic 'Demand prediction'
To see the other types of publications on this topic, follow the link: Demand prediction.
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Demand prediction.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
1
McElroy, Wade Allen. "Demand prediction modeling for utility vegetation management." Thesis, Massachusetts Institute of Technology, 2018. http://hdl.handle.net/1721.1/117973.
Full textAbstract:
Thesis: M.B.A., Massachusetts Institute of Technology, Sloan School of Management, in conjunction with the Leaders for Global Operations Program at MIT, 2018.Thesis: S.M., Massachusetts Institute of Technology, Department of Mechanical Engineering, in conjunction with the Leaders for Global Operations Program at MIT, 2018.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 63-64).
This thesis proposes a demand prediction model for utility vegetation management (VM) organizations. The primary uses of the model is to aid in the technology adoption process of Light Detection and Ranging (LiDAR) inspections, and overall system planning efforts. Utility asset management ensures vegetation clearance of electrical overhead powerlines to meet state and federal regulations, all in an effort to create the safest and most reliable electrical system for their customers. To meet compliance, the utility inspects and then prunes and/or removes trees within their entire service area on an annual basis. In recent years LiDAR technology has become more widely implemented in utilities to quickly and accurately inspect their service territory. VM programs encounter the dilemma of wanting to pursue LiDAR as a technology to improve their operations, but find it prudent, especially in the high risk and critical regulatory environment, to test the technology. The biggest problem during, and after, the testing is having a baseline of the expected number of tree units worked each year due to the intrinsic variability of tree growth. As such, double inspection and/or long pilot projects are conducted before there is full adoption of the technology. This thesis will address the prediction of circuit-level tree work forecasting through the development a model using statistical methods. The outcome of this model will be a reduced timeframe for complete adoption of LiDAR technology for utility vegetation programs. Additionally, the modeling effort provides the utility with insight into annual planning improvements. Lastly for later usage, the model will be a baseline for future individual tree growth models that include and leverage LiDAR data to provide a superior level of safety and reliability for utility customers.
by Wade Allen McElroy.
M.B.A.
S.M.
2
Zhou, Yang. "Multi-Source Large Scale Bike Demand Prediction." Thesis, University of North Texas, 2020. https://digital.library.unt.edu/ark:/67531/metadc1703413/.
Full textAbstract:
Current works of bike demand prediction mainly focus on cluster level and perform poorly on predicting demands of a single station. In the first task, we introduce a contextual based bike demand prediction model, which predicts bike demands for per station by combining spatio-temporal network and environment contexts synergistically. Furthermore, since people's movement information is an important factor, which influences the bike demands of each station. To have a better understanding of people's movements, we need to analyze the relationship between different places. In the second task, we propose an origin-destination model to learn place representations by using large scale movement data. Then based on the people's movement information, we incorporate the place embedding into our bike demand prediction model, which is built by using multi-source large scale datasets: New York Citi bike data, New York taxi trip records, and New York POI data. Finally, as deep learning methods have been successfully applied to many fields such as image recognition and natural language processing, it inspires us to incorporate the complex deep learning method into the bike demand prediction problem. So in this task, we propose a deep spatial-temporal (DST) model, which contains three major components: spatial dependencies, temporal dependencies, and external influence. Experiments on the NYC Citi Bike system show the effectiveness and efficiency of our model when compared with the state-of-the-art methods.
3
Sun, Rui S. M. Massachusetts Institute of Technology. "Analytics for hotels : demand prediction and decision optimization." Thesis, Massachusetts Institute of Technology, 2017. http://hdl.handle.net/1721.1/111438.
Full textAbstract:
Thesis: S.M. in Transportation, Massachusetts Institute of Technology, Department of Civil and Environmental Engineering, 2017.Thesis: S.M., Massachusetts Institute of Technology, Department of Electrical Engineering and Computer Science, 2017.
Cataloged from PDF version of thesis.
Includes bibliographical references (pages 69-71).
The thesis presents the work with a hotel company, as an example of how machine learning techniques can be applied to improve demand predictions and help a hotel property to make better decisions on its pricing and capacity allocation strategies. To solve the decision optimization problem, we first build a random forest model to predict demand under given prices, and then plug the predictions into a mixed integer program to optimize the prices and capacity allocation decisions. We present in the numerical results that our demand forecast model can provide accurate demand predictions, and with optimized decisions, the hotel is able to obtain a significant increase in revenue compared to its historical policies.
by Rui Sun.
S.M. in Transportation
S.M.
4
Svensk, Gustav. "TDNet : A Generative Model for Taxi Demand Prediction." Thesis, Linköpings universitet, Programvara och system, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-158514.
Full textAbstract:
Supplying the right amount of taxis in the right place at the right time is very important for taxi companies. In this paper, the machine learning model Taxi Demand Net (TDNet) is presented which predicts short-term taxi demand in different zones of a city. It is based on WaveNet which is a causal dilated convolutional neural net for time-series generation. TDNet uses historical demand from the last years and transforms features such as time of day, day of week and day of month into 26-hour taxi demand forecasts for all zones in a city. It has been applied to one city in northern Europe and one in South America. In northern europe, an error of one taxi or less per hour per zone was achieved in 64% of the cases, in South America the number was 40%. In both cities, it beat the SARIMA and stacked ensemble benchmarks. This performance has been achieved by tuning the hyperparameters with a Bayesian optimization algorithm. Additionally, weather and holiday features were added as input features in the northern European city and they did not improve the accuracy of TDNet.
5
Lu, Hongwei Marketing Australian School of Business UNSW. "Small area market demand prediction in the automobile industry." Publisher:University of New South Wales. Marketing, 2008. http://handle.unsw.edu.au/1959.4/43027.
Full textAbstract:
The general aim of this research is to investigate approaches to: improve small area market demand (i.e. SAMD) prediction accuracy for the purchase of automobiles at the level of each Census Collection District (i.e. CCD); and enhance understanding of meso-level marketing phenomena (i.e. geographically aggregated phenomena) relating to SAMD. Given the importance of SAMD prediction, and the limitations posed by current methods, four research questions are addressed: What are the key challenges in meso-level SAMD prediction? What variables affect SAMD prediction? What techniques can be used to improve SAMD prediction? What is the value of integrating these techniques to improve SAMD prediction? To answer these questions, possible solutions from two broad areas are examined: spatial analysis and data mining. The research is divided into two main studies. In the first study, a seven-step modelling process is developed for SAMD prediction. Several sets of models are analysed to examine the modelling techniques effectiveness in improving the accuracy of SAMD prediction. The second study involves two cases to: 1) explore the integration of these techniques and their advantages in SAMD prediction; and 2) gain insights into spatial marketing issues. The case study of Peugeot in the Sydney metropolitan area shows that urbanisation and geo-marketing factors can have a more important role in SAMD prediction than socio-demographic factors. Furthermore, results show that modelling spatial effects is the most important aspect of this prediction exercise. The value of the integration of techniques is in compensating for the weaknesses of conventional techniques, and in providing complementary and supplementary information for meso-level marketing analyses. Substantively, significant spatial variation and continuous patterns are found with the influence of key studied variables. The substantive implications of these findings have a bearing on both academic and managerial understanding. Also, the innovative methods (e.g. the SAMD modelling process and the model cube based technique comparison) developed from this research make significant contributions to marketing research methodology.
6
Lönnbark, Carl. "On Risk Prediction." Doctoral thesis, Umeå universitet, Nationalekonomi, 2009. http://urn.kb.se/resolve?urn=urn:nbn:se:umu:diva-22200.
Full textAbstract:
This thesis comprises four papers concerning risk prediction. Paper [I] suggests a nonlinear and multivariate time series model framework that enables the study of simultaneity in returns and in volatilities, as well as asymmetric effects arising from shocks. Using daily data 2000-2006 for the Baltic state stock exchanges and that of Moscow we find recursive structures with Riga directly depending in returns on Tallinn and Vilnius, and Tallinn on Vilnius. For volatilities both Riga and Vilnius depend on Tallinn. In addition, we find evidence of asymmetric effects of shocks arising in Moscow and in the Baltic states on both returns and volatilities. Paper [II] argues that the estimation error in Value at Risk predictors gives rise to underestimation of portfolio risk. A simple correction is proposed and in an empirical illustration it is found to be economically relevant. Paper [III] studies some approximation approaches to computing the Value at Risk and the Expected Shortfall for multiple period asset re- turns. Based on the result of a simulation experiment we conclude that among the approaches studied the one based on assuming a skewed t dis- tribution for the multiple period returns and that based on simulations were the best. We also found that the uncertainty due to the estimation error can be quite accurately estimated employing the delta method. In an empirical illustration we computed five day Value at Risk's for the S&P 500 index. The approaches performed about equally well. Paper [IV] argues that the practise used in the valuation of the port- folio is important for the calculation of the Value at Risk. In particular, when liquidating a large portfolio the seller may not face horizontal de- mandcurves. We propose a partially new approach for incorporating this fact in the Value at Risk and in an empirical illustration we compare it to a competing approach. We find substantial differences.
7
Bernhardsson, Viktor, and Rasmus Ringdahl. "Real time highway traffic prediction based on dynamic demand modeling." Thesis, Linköpings universitet, Kommunikations- och transportsystem, 2014. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-112094.
Full textAbstract:
Traffic problems caused by congestion are increasing in cities all over the world. As a traffic management tool traffic predictions can be used in order to make prevention actions against traffic congestion. There is one software for traffic state estimations called Mobile Millennium Stockholm (MMS) that are a part of a project for estimate real-time traffic information.In this thesis a framework for running traffic predictions in the MMS software have been implemented and tested on a stretch north of Stockholm. The thesis is focusing on the implementation and evaluation of traffic prediction by running a cell transmission model (CTM) forward in time.This method gives reliable predictions for a prediction horizon of up to 5 minutes. In order to improve the results for traffic predictions, a framework for dynamic inputs of demand and sink capacity has been implemented in the MMS system. The third part of the master thesis presents a model which adjusts the split ratios in a macroscopic traffic model based on driver behavior during congestion.
8
Jones, Simon Andrew. "Prediction of demand for emergency care in an acute hospital." Thesis, Kingston University, 2005. http://eprints.kingston.ac.uk/20739/.
Full textAbstract:
This thesis describes some models that attempt to forecast the number of occupied beds due to emergency admissions each day in an acute general hospital. Hospital bed managers have two conflicting demands: they must not only ensure that at all times they have sufficient empty beds to cope with possible emergency admissions but they must fill as many empty beds as possible with people on the waiting list. This model is important as it could help balance these two conflicting demands. The research is based on data from a district general and a postgraduate teaching hospital in South East London. Several tests indicate that emergency bed occupancy may have a nonlinear underlying data generating process. Therefore, both linear models and nonlinear models have been fitted to the data. At horizons up to 14 days, it was found that there was no statistically significant difference in the errors from the linear and nonlinear models. However at the 35 day forecast horizon the linear model gives the best forecast and tests indicate errors from this model are within 4% of mean occupancy. It is noted that a Markov Switching model gave very good forecasts of up to 4 days into the future. A search of the literature found no previous research that tested emergency bed occupancy for nonlinearities. The thesis ends with a gravity model to predict the change in number of Accident and Emergency (A&E) attendances following the relocation of an A&E Department in South East London.
9
Shen, Ni. "Prediction of International Flight Operations at U.S. Airports." Thesis, Virginia Tech, 2006. http://hdl.handle.net/10919/35687.
Full textAbstract:
This report presents a top-down methodology to forecast annual international flight operations at sixty-six U.S. airports, whose combined operations accounted for 99.8% of the total international passenger flight operations in National Airspace System (NAS) in 2004. The forecast of international flight operations at each airport is derived from the combination of passenger flight operations at the airport to ten World Regions. The regions include: Europe, Asia, Africa, South America, Mexico, Canada, Caribbean and Central America, Middle East, Oceania and U.S. International. In the forecast, a "top-down" methodology is applied in three steps. In the fist step, individual linear regression models are developed to forecast the total annual international passenger enplanements from the U.S. to each of nine World Regions. The resulting regression models are statistically valid and have parameters that are credible in terms of signs and magnitude. In the second step, the forecasted passenger enplanements are distributed among international airports in the U.S. using individual airport market share factors. The airport market share analysis conducted in this step concludes that the airline business is the critical factor explaining the changes associated with airport market share. In the third and final step, the international passenger enplanements at each airport are converted to flight operations required for transporting the passengers. In this process, average load factor and average seats per aircraft are used.
The model has been integrated into the Transportation Systems Analysis Model (TSAM), a comprehensive intercity transportation planning tool. Through a simple graphic user interface implemented in the TSAM model, the user can test different future scenarios by defining a series of scaling factors for GDP, load factor and average seats per aircraft. The default values for the latter two variables are predefined in the model using 2004 historical data derived from Department of Transportation T100 international segment data.
Master of Science
10
Paul, Udita. "Efficient access network selection and data demand prediction for 5G systems." Master's thesis, University of Cape Town, 2018. http://hdl.handle.net/11427/29729.
Full textAbstract:
The massive proliferation of sophisticated mobile terminals with advanced capabilities have led to an enormous surge in the demand for mobile broadband data. Also, the recent popularity of bandwidth intensive applications such as Netflix and YouTube has contributed to this demand for the wireless resources. In order to cope with this massive demand, fifth generation (5G) of wireless network is on the verge of deployment. This new generation of the wireless networks would pose different challenges for both subscribers and service providers, and the challenges need to be carefully addressed. Due to the diverse nature of the subscribers of mobile broadband, one network element is inadequate to meet the imposed requirements. Subscribers vary in terms of their usage of wireless resources as well as their preferred content. Deployment of the 5G systems promises the introduction of multiple tiers of heterogeneous networks within its architecture. This means radio access technologies (RATs) of various kinds (2G, 3G, 4G, 5G and Wi-Fi) would have to co-exist and aim to bridge the gap between the supply and demand for data. Subscribers, equipped with multi-mode or multi homing mobile terminals, can connect to one or more RATs to receive the required services. They also often run multiple applications simultaneously and as such, it must be ensured that the best access technology is assigned to a particular subscriber to maintain quality of experience and service. As such, an algorithm need to be devised that selects the best network to provide ubiquitous coverage to different types of users, running various kinds of applications, under dynamic network conditions. The network and infrastructure providers, on the other hand, face the need to meet up with the demand for data that the subscribers in different coverage regions require. In the 5G system, traditional proprietary hardware performing dedicated network functions such as packet gateway and service gateway would be replaced by softwarized virtual network functions (VNFs). These VNFs would need to be hosted in the data centres and would require computational power to process the subscribers’ traffic originating in an area. Therefore, data centres are set to play a key role in the provisioning of service in 5G systems. However, before establishing a data centre in a region, the traffic profile of that region need to be carefully studied to determine the optimal position and dimension of the facility. Furthermore, as cellular traffic differs depending on the time of the day, accurate prediction models are required to forecast future traffic demand to ensure dynamic and proper utilization of resources. This thesis aims to propose solutions to address these problems that subscribers and infrastructure providers face. Firstly, an algorithm is proposed to select the best access network for a subscriber running single or group of applications. Deviating from the existing access selection schemes in the literature, which consider the RAT-selection problem in an environment where accurate information is always available, the proposed algorithm models the problem in a completely fuzzy environment. As wireless networks are highly dynamic systems that are not only very unpredictable but also susceptible to sudden changes (for example malfunction of a particular RAT rendering it unusable), fuzzy systems are most adept in representing them. In the proposed algorithm, a new branch of fuzzy logic, Intuitionistic Fuzzy (IF) logic, is used with a popular multi-criteria decision making (MCDM) algorithm -Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), to formulate a network selection problem. The IF-TOPSIS scheme is designed to accurately take in various parameters such as network conditions, different number of applications and user preferences to select the ideal network for different types of subscribers. The second part of this thesis aims to solve the problem associated with establishment of data centre and utilization of its resources. As the cellular traffic exhibit strong spatial and temporal dependencies, it becomes necessary to analyse the traffic before establishing an infrastructure like a data centre. Existing literature do not consider real world traffic while determining the best location and dimension of 5G data centres. In this thesis, a real world traffic data set is first analysed to understand the variations that are present in different regions within a city. Based on the traffic analysis, the ideal placement of the data centre is formulated as a facility location problem and solved using the Weiszfeld’s algorithm. Additionally, based on the traffic analysis, the optimal dimensions of the data centre in different regions are heuristically obtained. Finally, machine learning algorithms are employed to obtain future traffic demand values to aid dynamic allocation of data centre resources. Simulation results are presented to show the effectiveness of the proposed schemes.
11
Li, Yapeng. "Dynamic energy demand prediction and related control system for UK households." Thesis, University of Newcastle upon Tyne, 2015. http://hdl.handle.net/10443/2824.
Full textAbstract:
Domestic energy consumption is not only based on the type of appliances, weather conditions, and house type; it is also highly depended on related occupancy profiles. In order to manage and optimise energy generation and the effective use of energy storage, it is important to be able to accurately predict energy demand in advance. However, high-resolution (like below 1-min) occupancy profiles for domestic UK households are not ideally possible to be recorded or measured in nature. Therefore, an alternative approach to transfer particular electricity load to the number of active occupancy during selected time interval is identified by analysing the average electricity consumption of occupancy in this study. Real load data analysis for three type of participated UK households is presented throughout the year. Then the seasonal synthetic high-resolution (30s) occupancy patterns for each household are generated independently. Weekday occupancy profiles are collected seasonally and used in a Markov-Chain model to produce particular occupancy daily activity sequence for each household. A stochastic model by using Markov-Chain Monte Carlo is presented to randomly generate high-resolution occupancy profiles in dynamic. Then the predicted electricity loads are produced by mapping occupancy profiles to average electricity consumption. By validating the predicted results, it is found that maximum of sub-hourly aggregate result can mostly cover the measured demand in advance. Therefore, it is set the sub-hourly electricity demand boundary independently for each household during weekday throughout the year. Heat demand for each household is simulated in sub-hourly resolution by using DesignBuilder with EnergyPlus throughout the year. Thus, sub-hourly energy demand of each household is applied in the control system of Bio-fuel Micro Trigeneration with Hybrid Electrical Energy Storage. The control system is designed and implemented by using Siemens software STEP-7 S-300 and WinCC. In addition, the predicted energy demands are utilized into the optimization of the control system. The comparison of optimized and general control strategies shows that optimized strategies by applying prescient sub-hourly energy demand can improve system efficiency significantly.
12
Lindsey, Matthew Douglas. "Reliable Prediction Intervals and Bayesian Estimation for Demand Rates of Slow-Moving Inventory." Thesis, University of North Texas, 2007. https://digital.library.unt.edu/ark:/67531/metadc3946/.
Full textAbstract:
Application of multisource feedback (MSF) increased dramatically and became widespread globally in the past two decades, but there was little conceptual work regarding self-other agreement and few empirical studies investigated self-other agreement in other cultural settings. This study developed a new conceptual framework of self-other agreement and used three samples to illustrate how national culture affected self-other agreement. These three samples included 428 participants from China, 818 participants from the US, and 871 participants from globally dispersed teams (GDTs). An EQS procedure and a polynomial regression procedure were used to examine whether the covariance matrices were equal across samples and whether the relationships between self-other agreement and performance would be different across cultures, respectively. The results indicated MSF could be applied to China and GDTs, but the pattern of relationships between self-other agreement and performance was different across samples, suggesting that the results found in the U.S. sample were the exception rather than rule. Demographics also affected self-other agreement disparately across perspectives and cultures, indicating self-concept was susceptible to cultural influences. The proposed framework only received partial support but showed great promise to guide future studies. This study contributed to the literature by: (a) developing a new framework of self-other agreement that could be used to study various contextual factors; (b) examining the relationship between self-other agreement and performance in three vastly different samples; (c) providing some important insights about consensus between raters and self-other agreement; (d) offering some practical guidelines regarding how to apply MSF to other cultures more effectively.
13
Pan, Song. "On demand DBS for Parkinson's Disease : tremor prediction using artificial neural networks." Thesis, University of Reading, 2012. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.567590.
Full textAbstract:
In this thesis results are presented which relate to using artificial neural networks to predict the onset of Parkinson's disease tremors in human subjects. Data for the networks was obtained from implanted deep brain electrodes in human subjects. A tuned artificial neural network was shown to be able to identify the pattern of the onset tremor from these real time recordings. Parkinson's disease (PO) is one disease in a group of conditions called movement disorders. One of the primary symptoms of Parkinson's disease is tremor, and in the extreme case, the patient can suffer loss of physical movement. There are two major types of treatment for PO currently available, namely chemical treatment (Levodopa) and surgical implants (Deep Brain Stimulation). Deep Brain Stimulation (DBS) has been widely accepted as an efficient treatment for PO over the past decade. Despite the high cost of surgical operation, deep brain stimulation has become a widely accepted alternative choice (if not the only) to medical treatment such as Levodopa for patients. In this work, number of methods have been applied on exploring the possibility of determining PO tremor onset from patient's brain signal, in particular using combination of artificial neural networks (ANN) and advanced signal processing algorithms. The result of this work could eventually lead to design a deep brain stimulation device with the ability to react on different brain activities, for example, start stimulation just before Parkinson's disease tremor onset. The benefits of such smart device are pre-Iong DBS battery life and reduce stimulation interference on normal brain functions.
14
Lindsey, Matthew Douglas Pavur Robert J. "Reliable prediction intervals and Bayesian estimation for demand rates of slow-moving inventory." [Denton, Tex.] : University of North Texas, 2007. http://digital.library.unt.edu/permalink/meta-dc-3946.
Full text
15
Stojanovski, Filip. "Churn Prediction using Sequential Activity Patterns in an On-Demand Music Streaming Service." Thesis, KTH, Skolan för informations- och kommunikationsteknik (ICT), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-228226.
Full textAbstract:
In data-driven companies, churn analysis aims to make use of novel machine learning and data mining techniques for the purpose of better understanding of the customers. The most common approach is to engineer a vast number of features describing users, products, services, and actions, which are then used to infer knowledge by means of machine learning and data mining. However, one aspect is typically neglected since it appears more difficult to model and utilize, and that is the time. This work presents the modeling of user activity on a music streaming service in the form of sequential temporally-dependent data, which serves to explore the advantages of detecting churning users by means of basic or long short-term memory recurrent neural networks. The performance and complexity are compared against non-sequential models using the same data. The conclusion reached is that even though recurrent networks bring no improvement of the module of a churn prediction model based on activity data, that data presented in sequential form does.Inom datadrivna företag används churn-analys i formv av maskininlärning och datautvinningsmetoder i syfte att bättre förstå kunder som väljer att terminera deras relation till företag. Det vanligaste angreppssättet är att ta fram ett antal attribut som kännetecknar användare, produkter, tjänster och handlingar, vilka sedan används för att generera kunskap med hjälp av maskinlärning och informationsutvinning. En aspekt som dock vanligtvis försummas är tid, då den visar sig vara svår att modellera och använda. Detta arbete presenterar modelleringen av användaraktivitet hos en musikstreamingtjänst i form av sekventiell temporärt beroende data, i syfte att utforska fördelarna med att detektera kunder som väljer att terminera sitt medlemskap med hjälp av long short-term recurrent neurala nätverk. Prestanda och komplexitet jämförs med icke-sekventiella modeller med samma data. Arbetets slutsats är att trots att recurrent neurala nätverk inte resulterar i en förbättrad churn-predikteringsmodul baserad på aktivitetsdata, så ger data presenterad i sekventiell form en förbättring.
16
Zhang, Huimin. "User Behavior Analysis and Prediction Methods for Large-scale Video-on- demand System." Thesis, Uppsala universitet, Institutionen för informationsteknologi, 2015. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-263261.
Full textAbstract:
Video-on-demand (VOD) systems are some of the best-known examples of 'next-generation' Internet applications. With their growing popularity, huge amount of video content imposes a heavy burden on Internet traffic which, in turns, influences the user experience of the systems. Predicting and pre- fetching relevant content before user requests is one of the popular methods used to reduce the start-up delay. In this paper, a typical VOD system is characterized and user's watching behavior is analyzed. Based on the characterization, two pre- fetching approaches based on user behavior are investigated. One is to prediction relevant content based on access history. The other is prediction based on user-clustering. The results clearly indicate the value of pre-fetching approaches for VOD systems and lead to the discussions on future work for further improvement.
17
Sun, Wenzhe. "Bus Bunching Prediction and Transit Route Demand Estimation Using Automatic Vehicle Location Data." Kyoto University, 2020. http://hdl.handle.net/2433/253498.
Full text
18
Hast, Matteus. "Evaluation of machine learning algorithms for customer demand prediction of in-flight meals." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-255020.
Full textAbstract:
This study aims to evaluate multiple Machine Learning Algorithms (MLAs) for estimating the customer demand of in-flight meals. As a result of the review of related works, four MLAs were selected, namely Linear Regression (LR), Support Vector Regression (SVR), Extreme Gradient Boosting (XGBoost) and a Multilayer Perceptron Neural Network (MLP). The study investigates which MLA is best suited for the problem at hand and which features are most influential for customer demand prediction of in-flight meals. Focus is put on finding applicable MLAs and on evaluating, comparing and tweaking the parameters of the MLAs to further optimise the selected models. The available data set comes from a single airline company and consists mainly of flights with a short to medium long flight duration time.The results show that the four evaluated models, LR, SVR, XGBoost and MLP performs with no significant difference against one another and are comparable in their performance in regard to estimation accuracy with results close to each other’s. However, the SVR model underperforms in regard to model fitting and prediction time in comparison towards the remaining three models. Furthermore, the most important feature for customer demand prediction of in-flight meals is the scheduled flight duration time.Syftet med den här studien är att utvärdera ett flertal maskininlärningsalgoritmer för prediktering av konsumentefterfrågan för måltider under flygning. Undersökningen över tidigare arbeten utförda i liknande fält resulterade i att fyra maskininlärningsalgoritmer blev valda, nämligen linjär regression, stödvektormaskin för regression, Extreme Gradient Boosting och ett flerlagersperceptron-neuronnät. Studien utforskar vilken maskininlärningsalgoritm som är bäst anpassad för att prediktera problemet samt vilka egenskaper i datat som är mest inflytesrika när det handlar om att prediktera konsumentefterfrågan av måltider under flygning. Fokus ligger på att finna applicerbara maskininlärningsalgoritmer och på att utvärdera, jämföra samt på att justera parametrarna i syfte till att optimera modellerna. Den tillgängliga datan härstammar från ett enstaka flygbolag och består mestadels av korta och mediumlånga flyg.Resultatet påvisar att de fyra modellerna, linjär regression, en stödvektormaskin för regression, Extreme Gradient Boosting och ett flerlagersperceptron-neuronnät presterar utan någon signifikant skillnad gentemot varandra och är jämförbara i deras prestation i avseende till predikteringprecision med liknande resultat. I avseende till modellanpassningsoch predikteringstid underpresterar dock stödvektormaskinen avsevärt i jämförelse med de resterande tre modellerna. Resultatet visar även att den viktigaste egenskapen i datat för prediktering av konsumentefterfrågan av måltider under flygning är den schemalagda flygtiden.
19
Farzana, Fatema Hoque. "Estimation and Prediction of Mobility and Reliability Measures Using Different Modeling Techniques." FIU Digital Commons, 2018. https://digitalcommons.fiu.edu/etd/3880.
Full textAbstract:
The goal of this study is to investigate the predictive ability of less data intensive but widely accepted methods to estimate mobility and reliability measures. Mobility is a relatively mature concept in the traffic engineering field. Therefore, many mobility measure estimation methods are already available and widely accepted among practitioners and researchers. However, each method has their inherent weakness, particularly when they are applied and compared with real-world data. For instances, Bureau of Public Roads (BPR) Curves are very popular in static route choice assignment, as part of demand forecasting models, but it is often criticized for underperforming in congested traffic conditions where demand exceeds capacity. This study applied five mobility estimation methods (BPR Curve, Akcelic Function, Florida State University (FSU) Regression Model, Queuing Theory, and Highway Capacity Manual (HCM) Facility Procedures) for different facility types (i.e. Freeway and Arterial) and time periods (AM Peak, Mid-Day, PM Peak). The study findings indicate that the methods were able to accurately predict mobility measures (e.g. speed and travel time) on freeways, particularly when there was no congestion and the volume was less than the capacity. In the presence of congestion, none of the mobility estimation methods predicted mobility measures closer to the real-world measure. However, compared with the other prediction models, the HCM procedure method was able to predict mobility measures better. On arterials, the mobility measure predictions were not close to the real-world measurements, not even in the uncongested periods (i.e. AM Peak and Mid-Day). However, the predictions are relatively better in the AM and Mid-Day periods that have lower volume/capacity ration compared to the PM Peak period.
To estimate reliability measures, the study applied three products from the Second Strategic Highway Research Program (SHRP2) projects (Project Number L03, L07, and C11) to estimate three reliability measures; the 80th percentile travel time index, 90th percentile travel time index, and 95th percentile travel time index. A major distinction between mobility estimation process and reliability estimation process lies in the fact that mobility can be estimated for any particular day, but reliability estimation requires a full year of data. Inclusion of incident days and weather condition are another important consideration for reliability measurements. The study found that SHRP2 products predicted reliability measures reasonably well for freeways for all time periods (except C11 in the PM Peak). On arterials, the reliability predictions were not close to the real-world measure, although the differences were not as drastic as seen in the case of arterial mobility measures.
20
Eriksson, Niclas. "Predicting demand in districtheating systems : A neural network approach." Thesis, Uppsala universitet, Avdelningen för beräkningsvetenskap, 2012. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-175082.
Full textAbstract:
To run a district heating system as efficiently as possible correct unit-commitmentdecisions has to be made and in order to make those decisions a good forecast ofheat demand for the coming planning period is necessary. With a high quality forecastthe need for backup power and the risk for a too high production are lowered. Thisthesis takes a neural network approach to load forecasting and aims to provide asimple, yet powerful, tool that can provide accurate load forecasts from existingproduction data without the need for extensive model building.The developed software is tested using real life data from two co-generation plantsand the conclusion is that when the quality of the raw data is good, the software canproduce very good forecasting results.
21
Ghareeb, Ahmed. "Data mining for University of Dayton campus buildings to predict future demand." University of Dayton / OhioLINK, 2017. http://rave.ohiolink.edu/etdc/view?acc_num=dayton1490472227466522.
Full text
22
Kerakos, Emil, Oscar Lindgren, and Vladislav Tolstoy. "Machine Learning for Ambulance Demand Prediction in Stockholm County : Towards efficient and equitableDynamic Deployment Systems." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-282420.
Full textAbstract:
Pre-hospital care is a widely discussed subject with many actors working on figuring out what factors determine the outcome for the patient and how those factors can be affected. One factor believed to have a major impact on patient outcome is ambulance response time. A proposed way to improve response time is dynamic deployment systems. These systems require detailed predictions of spatio-temporal ambulance demand in order to function effectively. The purpose of the study is to explore the possibility of using machine learning to build a high-resolution predictor that dynamic deployment systems can use to reduce response time. In this paper we first try out unsupervised machine learning algorithms to dividing Stockholm County into small subregions (clusters) over which predictions can be made. Then, based on the best cluster-structure obtained, we train and evaluate a logistic regression model an to make probabilistic predictions of ambulance demand over these clusters. We compare it to a baseline model and although the logistic regression model outperforms the baseline in total, it is worse at predicting when dispatches actually happens. Either way, we see that risk-terrain data and historic dispatches data seem to be useful for predicting ambulance demand. In the end of this paper we evaluate how suitable today’s key performance indicators (KPI:s) are for Emergency Medical Systems(EMS) implementing dynamic deployment. We find that these systems likely entail a need for updated KPI:s for measuring effectiveness and equity.Den prehospitala vården består av många aktörer och många faktorer som påverkar hälsoutfallen hos de behövande patienterna. En central faktor är ambulansens utryckningstid där ett föreslaget sätt att minska utryckningstiderna är via dynamisk positionering (eng: dynamic deployment). Dessa system kräver prediktioner om var och när ambulanser kommer behövas på hög detaljnivå geografiskt och tidsmässigt för att fungera effektivt. I denna studie utvärderar vi först så kallade oövervakade maskininlärningsalgoritmer för att dela in Stockholms Län i mindre områden och ser att områdena blir rimliga. Över dessa områden tränar vi logistisk regression för att göra sådana prediktioner i hög detaljnivå om det kommande behovet och jämför med en enkel model. Vi ser att även om logistisk regression ger högre prediktionskraft totalt än den enkla modellen är den sämre än den enkla modellen på att prediktera när behov faktiskt uppstår. Vi ser dock ändå att data om historiska utryckningar och riskterräng verkar vara användbara för att göra sådana prediktioner. Relaterat till detta analyserar och diskuterar vi även hur väl lämpade dagens prestationsmått är för att evaluera prehospitala vårdsystem som implementerar dynamisk positionering. Vi ser att dessa system troligen behöver justera sina huvudsakliga KPI:er för att mäta effektivitet och jämlikhet på ett bra sätt.
23
Xu, Yizheng. "Probabilistic estimation and prediction of the dynamic response of the demand at bulk supply points." Thesis, University of Manchester, 2015. https://www.research.manchester.ac.uk/portal/en/theses/probabilistic-estimation-and-prediction-of-the-dynamic-response-of-the-demand-at-bulk-supply-points(b9e427ec-7e5e-49a5-aec4-0f34032d71a9).html.
Full textAbstract:
The dynamic response of the demand is defined as the time-domain real and reactive power response to a voltage disturbance, and it represents the dynamic load characteristics. This thesis develops a methodology for probabilistic estimation and prediction of dynamic responses of the demand at bulk supply points. The main outcome of the research is being able to predict the contribution of different categories of loads to the total demand mix and their controllability without conducting detailed customer surveys or collecting smart meter data, and to predict the dynamic response of the demand without performing field tests. The prediction of the contributions of different load categories and their controllability and load characteristics in the near future (e.g., day ahead) plays an important role in system analysis and planning, especially in the short-term dispatch and control. However, the research related to this topic is missing in the publically available literature, and an approach needs to be developed to enable the prediction of the participation of different loads in total load mix, their controllability and the dynamic response of the demand. This research contributes to a number of areas, such as load forecasting, load disaggregation and load modelling. First, two load forecasting methodologies which have not been compared before are compared; and based on the results of comparison and considering the actual requirements in this research, a methodology is selected and used to predict both the real and reactive power. Second, a unique methodology for load disaggregation is developed. This methodology enables the estimation of the contributions of different load categories to the total demand mix and their controllability based on RMS measured voltage and real and reactive power. The confidence level of the estimation is also assessed. The methodology for disaggregation is integrated with the load forecasting tool to enable prediction of load compositions and dynamic responses of the demand. The prediction is validated with data collected from real UK power network. Finally, based on the prediction, an example of load shifting is used to demonstrate that different dynamic responses can be obtained based on the availability and redistribution of controllable devices and that load shifting decisions, i.e., demand side management actions, should be made based not only on the amount of demand to be shifted, but also on predicted responses before and after load shifting.
24
Karlsson, Niclas, and Zandra Karlsson. "Detecting Disruption: : an Ex-ante Study in the Automotive Industry." Thesis, Blekinge Tekniska Högskola, Institutionen för industriell ekonomi, 2016. http://urn.kb.se/resolve?urn=urn:nbn:se:bth-12834.
Full textAbstract:
In history there are numerous examples of strong market-leaders who have lost everything through the emergence of a new breakthrough technology which has replaced the existing one. That could be the reason why Christensen received such high attention when he presented his famous work about disruptive technologies in 1997. In his work, and in many following studies, several aspects of this phenomenon have been investigated. However, the key point for the market leaders, the ability of identifying a market disruption before it happens, ex ante, is still a field that has not reached a definedstate of the art. In this work one of Christensen's original ideas of disruption, driven by changes incustomer demand, is highlighted as a possible improvement for the ex ante methodology. In this thesis a selected existing holistic prediction model is extended explicitly with this aspect of need change. The purpose of this work is thus to evaluate if including the property of shifting customer needs in an existing holistic model would improve the ex ante prediction of disruption and lead to a simple, practical but yet rich model. With a literature review of the existing types of ex ante methods a fitting base model for the holistic approach to disruptive prediction is found. A second literature review is performed with the focus on disruption and its link to changes in need, as expressed by customer demand. This serves as a starting point for the extension of the base ex ante model into a methodology that look also upon the aspect of shifting customer demand. To validate and use the proposed extended ex ante model a qualitative approach is selected. It consists of two studies within the automotive industry. One is a history analysis of a known disruptive case to validate the extension, the entry of Japanese car manufacturers into the US market. One is a case-study of a present potentially disruptive case to apply the extended method as a genuine ex ante method for final evaluation at a post-disruption stage. It investigates the effects of electric vehicles on the Chinese automotive market. Through the analysis of the two studies the conclusion is reached that a qualitative improvement of the prediction has been obtained. Additionally it is shown that the analysis of customer need change can provide an increased understanding of the underlying drivers of the disruption.
25
Ahmed, Kishwar. "Energy Demand Response for High-Performance Computing Systems." FIU Digital Commons, 2018. https://digitalcommons.fiu.edu/etd/3569.
Full textAbstract:
The growing computational demand of scientific applications has greatly motivated the development of large-scale high-performance computing (HPC) systems in the past decade. To accommodate the increasing demand of applications, HPC systems have been going through dramatic architectural changes (e.g., introduction of many-core and multi-core systems, rapid growth of complex interconnection network for efficient communication between thousands of nodes), as well as significant increase in size (e.g., modern supercomputers consist of hundreds of thousands of nodes). With such changes in architecture and size, the energy consumption by these systems has increased significantly. With the advent of exascale supercomputers in the next few years, power consumption of the HPC systems will surely increase; some systems may even consume hundreds of megawatts of electricity. Demand response programs are designed to help the energy service providers to stabilize the power system by reducing the energy consumption of participating systems during the time periods of high demand power usage or temporary shortage in power supply.
This dissertation focuses on developing energy-efficient demand-response models and algorithms to enable HPC system's demand response participation. In the first part, we present interconnection network models for performance prediction of large-scale HPC applications. They are based on interconnected topologies widely used in HPC systems: dragonfly, torus, and fat-tree. Our interconnect models are fully integrated with an implementation of message-passing interface (MPI) that can mimic most of its functions with packet-level accuracy. Extensive experiments show that our integrated models provide good accuracy for predicting the network behavior, while at the same time allowing for good parallel scaling performance. In the second part, we present an energy-efficient demand-response model to reduce HPC systems' energy consumption during demand response periods. We propose HPC job scheduling and resource provisioning schemes to enable HPC system's emergency demand response participation. In the final part, we propose an economic demand-response model to allow both HPC operator and HPC users to jointly reduce HPC system's energy cost. Our proposed model allows the participation of HPC systems in economic demand-response programs through a contract-based rewarding scheme that can incentivize HPC users to participate in demand response.
26
Школа, Вікторія Юріївна, Виктория Юрьевна Школа, and Viktoriia Yuriivna Shkola. "Прогнозування попиту на інновації для формування стратегії розвитку держави." Thesis, Видавництво СумДУ, 2006. http://essuir.sumdu.edu.ua/handle/123456789/3811.
Full text
27
Zhou, Xuesong. "Dynamic origin-destination demand estimation and prediction for off-line and on-line dynamic traffic assignment operation." College Park, Md. : University of Maryland, 2004. http://hdl.handle.net/1903/1819.
Full textAbstract:
Thesis (Ph. D.) -- University of Maryland, College Park, 2004.Thesis research directed by: Civil Engineering. Title from t.p. of PDF. Includes bibliographical references. Published by UMI Dissertation Services, Ann Arbor, Mich. Also available in paper.
28
Candela, Garza Eduardo. "Revenue optimization for a hotel property with different market segments : demand prediction, price selection and capacity allocation." Thesis, Massachusetts Institute of Technology, 2017. http://hdl.handle.net/1721.1/113433.
Full textAbstract:
Thesis: S.M., Massachusetts Institute of Technology, Sloan School of Management, Operations Research Center, 2017.This electronic version was submitted by the student author. The certified thesis is available in the Institute Archives and Special Collections.
Cataloged from student-submitted PDF version of thesis.
Includes bibliographical references (pages 53-55).
We present our work with a hotel company as an example of how machine learning techniques can be used to improve the demand predictions of a hotel property, as well as its pricing and capacity allocation decisions. First, we build a price-sensitive random forest model to predict the number of daily bookings for each customer market segment. We feed these predictions into a mixed integer linear program (MILP) to optimize prices and capacity allocations at the same time. We prove that the MILP can be equivalently solved as a linear program, and then show that it produces upper and lower bounds for the expected revenue maximization Dynamic Program (DP), and that the gap between the bounds depends on the probabilistic distribution of the demand. Thus, for high prediction accuracies, the optimal value of the DP can be closely approximated by the MILP solution. Finally, numerical results show that the optimized decisions are able to generate an increase in revenue compared to the historical policies, and that the fast running time achieved permits real time policy updates.
by Eduardo Candela Garza.
S.M.
29
Boulin, Juan Manuel. "Call center demand forecasting : improving sales calls prediction accuracy through the combination of statistical methods and judgmental forecast." Thesis, Massachusetts Institute of Technology, 2010. http://hdl.handle.net/1721.1/59159.
Full textAbstract:
Thesis (M.B.A.)--Massachusetts Institute of Technology, Sloan School of Management; and, (S.M.)--Massachusetts Institute of Technology, Engineering Systems Division; in conjunction with the Leaders for Global Operations Program at MIT, 2010.Cataloged from PDF version of thesis.
Includes bibliographical references (p. 79-81).
Call centers are important for developing and maintaining healthy relationships with customers. At Dell, call centers are also at the core of the company's renowned direct model. For sales call centers in particular, the impact of proper operations is reflected not only in long-term relationships with customers, but directly on sales and revenue. Adequate staffing and proper scheduling are key factors for providing an acceptable service level to customers. In order to staff call centers appropriately to satisfy demand while minimizing operating expenses, an accurate forecast of this demand (sales calls) is required. During fiscal year 2009, inaccuracies in consumer sales call volume forecasts translated into approximately $1.1M in unnecessary overtime expenses and $34.5M in lost revenue for Dell. This work evaluates different forecasting techniques and proposes a comprehensive model to predict sales call volume based on the combination of ARIMA models and judgmental forecasting. The proposed methodology improves the accuracy of weekly forecasted call volume from 23% to 46% and of daily volume from 27% to 41%. Further improvements are easily achievable through the adjustment and projection processes introduced herein that rely on contextual information and the expertise of the forecasting team.
by Juan Manuel Boulin.
S.M.
M.B.A.
30
Ghias, Nezhad Omran Nima. "Power grid planning for vehicular demand: forecasting and decentralized control." IEEE Transactions on Smart Grid, 2014. http://hdl.handle.net/1993/23891.
Full textAbstract:
Temporal and spatial distribution of incoming vehicular charging demand is a significant challenge for the future planning of power systems. In this thesis the vehicular loading is-sue is categorized into two classes of stationary and mobile; they are then addressed in two phases.
The mobile vehicular load is investigated first; a location-based forecasting algorithm for the charging demand of plug-in electric vehicles at potential off-home charging stations is proposed and implemented for real-world case-studies. The result of this part of the re-search is essential to realize the scale of fortification required for a power grid to handle vehicular charging demand at public charging stations.
In the second phase of the thesis, a novel decentralized control strategy for scheduling vehicular charging demand at residential distribution networks is developed. The per-formance of the proposed algorithm is then evaluated on a sample test feeder employing real-world driving data. The proposed charging scheduling algorithm will significantly postpone the necessity for upgrading the assets of the network while effectively fulfilling customers’ transportation requirements and preferences.October 2014
31
Martyr, Randall. "Optimal prediction games in local electricity markets." Thesis, University of Manchester, 2015. https://www.research.manchester.ac.uk/portal/en/theses/optimal-prediction-games-in-local-electricity-markets(976e566d-e942-444a-9ee0-df17f46188d4).html.
Full textAbstract:
Local electricity markets can be defined broadly as 'future electricity market designs involving domestic customers, demand-side response and energy storage'. Like current deregulated electricity markets, these localised derivations present specific stochastic optimisation problems in which the dynamic and random nature of the market is intertwined with the physical needs of its participants. Moreover, the types of contracts and constraints in this setting are such that 'games' naturally emerge between the agents. Advanced modelling techniques beyond classical mathematical finance are therefore key to their analysis. This thesis aims to study contracts in these local electricity markets using the mathematical theories of stochastic optimal control and games. Chapter 1 motivates the research, provides an overview of the electricity market in Great Britain, and summarises the content of this thesis. It introduces three problems which are studied later in the thesis: a simple control problem involving demand-side management for domestic customers, and two examples of games within local electricity markets, one of them involving energy storage. Chapter 2 then reviews the literature most relevant to the topics discussed in this work. Chapter 3 investigates how electric space heating loads can be made responsive to time varying prices in an electricity spot market. The problem is formulated mathematically within the framework of deterministic optimal control, and is analysed using methods such as Pontryagin's Maximum Principle and Dynamic Programming. Numerical simulations are provided to illustrate how the control strategies perform on real market data. The problem of Chapter 3 is reformulated in Chapter 4 as one of optimal switching in discrete-time. A martingale approach is used to establish the existence of an optimal strategy in a very general setup, and also provides an algorithm for computing the value function and the optimal strategy. The theory is exemplified by a numerical example for the motivating problem. Chapter 5 then continues the study of finite horizon optimal switching problems, but in continuous time. It also uses martingale methods to prove the existence of an optimal strategy in a fairly general model. Chapter 6 introduces a mathematical model for a game contingent claim between an electricity supplier and generator described in the introduction. A theory for using optimal switching to solve such games is developed and subsequently evidenced by a numerical example. An optimal switching formulation of the aforementioned game contingent claim is provided for an abstract Markovian model of the electricity market. The final chapter studies a balancing services contract between an electricity transmission system operator (SO) and the owner of an electric energy storage device (battery operator or BO). The objectives of the SO and BO are combined in a non-zero sum stochastic differential game where one player (BO) uses a classic control with continuous effects, whereas the other player (SO) uses an impulse control (discontinuous effects). A verification theorem proving the existence of Nash equilibria in this game is obtained by recursion on the solutions to Hamilton-Jacobi-Bellman variational PDEs associated with non-zero sum controller-stopper games.
32
Sapp, James Christopher. "Electricity Demand Forecasting in a Changing Regional Context: The Application of the Multiple Perspective Concept to the Prediction Process." PDXScholar, 1987. https://pdxscholar.library.pdx.edu/open_access_etds/574.
Full textAbstract:
In 1982, the Bonneville Power Administration (BPA), a marketer of hydroelectric power in the Pacific Northwest, found itself in a new role which required it to acquire power resources needed to meet the demands of the region's utilities. In particular, it had to deal with the Washington Public Power Supply System's nuclear plant cost escalations. In response, BPA prepared its first independent regional power forecast. The forecast development process was intricate and multidimensional and involved a variety of interested parties. Application of the Multiple Perspective Concept uncovers strengths and weaknesses in this process by illuminating its technical, organizational and personal dimensions. Examination of the forecast from the technical perspective revealed an elaborate set of interlinked models used to develop baseline, high, and low forecasts. The organizational perspective revealed BPA to be in a transitional stage. Internally, ratemaking, forecasting, conservation, resource acquisition, and financial management swelled as new organizational functions. Interorganizationally, environmentalists, ratepayer groups, and the region's utilities all had strong interests in the decision regarding WPPSS plants. The personal perspective revealed that each of the Administrators heading BPA since the early 1980s defined the agency's approach to the resource planning problem differently, first as an engineering problem, then as a political problem, and, finally, as a business problem. Taken together, the Multiple Perspectives yielded the following conclusions about BPA's 1982 forecast. (1) BPA's range forecast constituted a major improvement over the point forecasts preceding it, but left important classes of uncertainty unexplored. (2) BPA's models were better suited to address rate and conservation issues important at the time of the 1982 forecast than their predecessors. The model of the national economy, however, remained a black box, potentially significant feedbacks were not represented, and the sheer size of the modeling system placed practical limits on its use. (3) A stronger method of dealing with forecast uncertainty is needed which utilizes a disaster-avoidance strategy and plans for high impact/low probability events. This method need not involve the use of large models, but should incorporate qualitative insights from persons normally outside the technical sphere.
33
Borges, Viviana Marli Nogueira de Aquino. "Acoplamento de um modelo de previsão de demanda de água a um modelo simulador em tempo real - estudo de caso: sistema adutor metropolitano de São Paulo." Universidade de São Paulo, 2003. http://www.teses.usp.br/teses/disponiveis/3/3147/tde-17092004-101640/.
Full textAbstract:
O presente trabalho propõe uma evolução metodológica na operação do Sistema Adutor Metropolitano de São Paulo, em tempo real. Foi implantado um modelo matemático, em tempo real, de previsão de consumo de água horário para uma melhoria na performance operacional. Descrevem-se vários procedimentos de sistema de controle operacional, desde manual até totalmente automático, em sistemas de abastecimento. O sistema de abastecimento de São Paulo é classificado neste contexto. Foi analisada a possibilidade de desenvolvimento da situação atual rumo a um controle mais eficiente, através do uso de um modelo de previsão de demanda de água. O estado da arte em modelos de previsão de consumo de água é apresentado através de uma revisão bibliográfica especifica. Foi desenvolvida uma interface entre um modelo de rede hidráulica e um modelo de previsão de demanda de água existente, ambos utilizando dados operacionais, obtidos em tempo real de um sistema de telemetria. A interface foi testada em um estudo de caso do Sistema Adutor de São Paulo. Com a utilização de um modelo de previsão, concluiu-se que é possível estabelecer regras operacionais mais eficientes. Essa eficiência é demonstrada pela redução do número de mudanças de posição de válvula e estado de bombas, bem como é observada a redução do custo de energia elétrica (reduzindo o bombeamento em horário de maior custo). Os benefícios obtidos do uso conjunto do modelo simulador hidráulico e do modelo de previsão de demanda não podem ser considerados como o ótimo global. Seria necessário dispor de um modelo de otimização (programação automática). De qualquer forma, foi concluído que o investimento na implementação desses dois modelos é extremamente atrativa.This work proposes a methodological evolution of a real time water distribution system operation applied to the Water Mains System of Metropolitan Region of Sao Paulo. It was settled a mathematical model in real time, to forecast hourly water consumptions, intending to increase operational performance. Several operational control procedures of water systems were described, since manual ones until total automatic ones. Sao Paulo system is classified into this concept. The possibility of development from the present status toward a more efficient control was analyzed, through the use of a water demand prediction model. State-of-art of water demand models is presented, through a specific literature review. An interface between a hydraulic network model and an existing water demand prediction model were developed both of them using operational data, obtained in real time by a telemetric system. The interface was tested in a case study of Sao Paulo Water Mains System. One concludes that through the use of the prediction model, it was possible to make more efficient operational schedules. This efficiency is demonstrated by the reduction in number of valve positions changes and in pump status changes, as well as a decrease in energy costs could be observed ( reducing pump operations in hours of more expensive costs). Benefits obtained by the conjunctive use of the hydraulic simulation model and the water demand prediction model can not be admitted as the global optimum. It would be necessary to make available an optimization model (automatic scheduler). However it was concluded that investment in these two models implementations is extremely attractive.
34
Delin, Sofia. "Site-specific nitrogen fertilization demand in relation to plant available soil nitrogen and water : potential for prediction based on soil characteristics /." Skara : Department of Soil Sciences, Swedish University of Agricultural Sciences, 2005. http://epsilon.slu.se/200506.pdf.
Full text
35
Kang, Ying. "Estimation and prediction of dynamic origin-destination (O-D) demand and system consistency control for real-time dynamic traffic assignment operation /." Digital version accessible at:, 1999. http://wwwlib.umi.com/cr/utexas/main.
Full text
36
Goutham, Mithun. "Machine learning based user activity prediction for smart homes." The Ohio State University, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=osu1595493258565743.
Full text
37
Vicente, Rosmeiry Vanzella. "Modelo de operação para centros de controle de sistemas de abastecimento de água: estudo de caso - Sistema Adutor Metropolitano de São Paulo." Universidade de São Paulo, 2005. http://www.teses.usp.br/teses/disponiveis/3/3147/tde-10042006-141721/.
Full textAbstract:
O presente trabalho propõe um modelo de operação sustentado por um sistema de suporte à decisão para operar a distribuição de água em tempo real atendendo a condições / restrições hidráulicas com o mínimo custo de energia elétrica. O atendimento às condições / restrições hidráulicas são avaliadas por um modelo simulador hidráulico previamente montado e calibrado. O conjunto de resultados avaliados pelo modelo de simulação hidráulica é analisado por um modelo de otimização proposto com solução de programação linear. As condições de operação em tempo real geram a necessidade de alimentação de informações operacionais automáticas a qualquer momento e com curto espaço de tempo menor que horário. Para uma operação otimizada, previamente analisada por um modelo de simulação hidráulica cria uma condição critérios para uma previsão do consumo a ser atendido nas próximas horas. Um refinamento desses critérios são utilizados em um modelo de previsão de demanda de água que prevê e checa seus resultados de forma dinâmica. O modelo de operação proposto cria uma interface entre todos esses sistemas. Essa interface é testada e avaliada a partir de um estudo de caso aplicado no Sistema Adutor Metropolitano de São Paulo. A eficiência do modelo de operação proposto é apresentada tendo como resultado uma redução no custo de energia elétrica.This assignment considers an operation model supported by decision support systems to operates the water supply systems in real time, considering the hydraulical conditions while achieving some performance goals, in this case, reducing electricity costs (minimization of pumping costs) the attempt of the hydraulic constraints are evaluated by an hydraulical simulator previously calibrated. The set of results are analyzed by an optimization model which uses a linear programming. The operation conditions in real time requires automatic feeding operational information shortly at any time (less than an hour) for an optimized operation, previously analyzed by a hydraulic simulation model with creates condition criteria of consumption within following hours. These criteria are refined according to a demand prediction model that dynamically previews and checks the consumption results. This proposed model creates an interface between all these systems. This interface is tested and evaluated according to a study of the São Paulo´s metropolitan area, Sistema Alto Tietê. The efficiency of this proposed model is presented having reductions in the electric energy costs.
38
Faber, Andreas D. [Verfasser], Stefan [Gutachter] Spinler, and Arnd [Gutachter] Huchzermeier. "Data analytics in supply chain planning : applications in intermittent demand forecasting, partial defection prediction and price discrimination / Andreas D. Faber ; Gutachter: Stefan Spinler, Arnd Huchzermeier." Vallendar : WHU - Otto Beisheim School of Management, 2021. http://d-nb.info/1240764359/34.
Full text
39
Jonsson, Estrid, and Sara Fredrikson. "An Investigation of How Well Random Forest Regression Can Predict Demand : Is Random Forest Regression better at predicting the sell-through of close to date products at different discount levels than a basic linear model?" Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-302025.
Full textAbstract:
Allt eftersom klimatkrisen fortskrider ökar engagemanget kring hållbarhet inom företag. Växthusgaser är ett av de största problemen och matsvinn har därför fått mycket uppmärksamhet sedan det utnämndes till den tredje största bidragaren till de globala utsläppen. För att minska sitt bidrag rabatterar många matbutiker produkter med kort bästföredatum, vilket kommit att kräva en förståelse för hur priskänslig efterfrågan på denna typ av produkt är. Prisoptimering görs vanligtvis med så kallade Generalized Linear Models men då efterfrågan är ett komplext koncept har maskininl ärningsmetoder börjat utmana de traditionella modellerna. En sådan metod är Random Forest Regression, och syftet med uppsatsen är att utreda ifall modellen är bättre på att estimera efterfrågan baserat på rabattnivå än en klassisk linjär modell. Vidare utreds det ifall ett tydligt linjärt samband existerar mellan rabattnivå och efterfrågan, samt ifall detta beror av produkttyp. Resultaten visar på att Random Forest tar bättre hänsyn till det komplexa samband som visade sig finnas, och i detta specifika fall presterar bättre. Vidare visade resultaten att det sammantaget inte finns något linjärt samband, men att vissa produktkategorier uppvisar svag linjäritet.As the climate crisis continues to evolve many companies focus their development on becoming more sustainable. With greenhouse gases being highlighted as the main problem, food waste has obtained a great deal of attention after being named the third largest contributor to global emissions. One way retailers have attempted to improve is through offering close-to-date produce at discount, hence decreasing levels of food being thrown away. To minimize waste the level of discount must be optimized, and as the products can be seen as flawed the known price-to-demand relation of the products may be insufficient. The optimization process historically involves generalized linear regression models, however demand is a complex concept influenced by many factors. This report investigates whether a Machine Learning model, Random Forest Regression, is better at estimating the demand of close-to-date products at different discount levels than a basic linear regression model. The discussion also includes an analysis on whether discounts always increase the will to buy and whether this depends on product type. The results show that Random Forest to a greater extent considers the many factors influencing demand and is superior as a predictor in this case. Furthermore it was concluded that there is generally not a clear linear relation however this does depend on product type as certain categories showed some linearity.
40
Souza, Daniel Morais de. "Comparação de abordagens econométricas alternativas para modelagem da demanda anual de eletricidade no Brasil nos segmentos residencial, industrial e comercial." Universidade Federal de Juiz de Fora (UFJF), 2018. https://repositorio.ufjf.br/jspui/handle/ufjf/6891.
Full textAbstract:
Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-06-14T13:10:24Z
No. of bitstreams: 1
danielmoraisdesouza.pdf: 1277495 bytes, checksum: 7c2517a5b98a6f70aa787d954cd0c84a (MD5)Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-06-27T15:04:32Z (GMT) No. of bitstreams: 1 danielmoraisdesouza.pdf: 1277495 bytes, checksum: 7c2517a5b98a6f70aa787d954cd0c84a (MD5)
Made available in DSpace on 2018-06-27T15:04:32Z (GMT). No. of bitstreams: 1 danielmoraisdesouza.pdf: 1277495 bytes, checksum: 7c2517a5b98a6f70aa787d954cd0c84a (MD5) Previous issue date: 2018-02-19
Eletricidade é um insumo de uso generalizado nas economias modernas, penetrando nas mais variadas atividades produtivas e de consumo na sociedade. No entanto, as dificuldades de armazenamento em larga escala dessa forma de energia fazem com que a eletricidade seja muito sensível às condições de oferta, a ponto de que problemas de abastecimento rapidamente se convertem em apagões. Dentre vários dispositivos implementados na re-estrutuação do setor elétrico brasileiro (SEB) ao longo dos últimos 17 anos, estão sistemas de previsão de médio e longo-prazos usados por parte dos agentes públicos e privados do setor para reduzir as incertezas dos processos de abastecimento e expansão. A ANEEL chegou a recomendar na NT 292/2008-SER o uso de três metodologias multivariadas alternativas nesses sistemas de previsão, a saber: modelos VAR e VCE, modelos autorregressivos com defasagens distribuídas (ARDL) e modelos estruturais a espaço de estados. A literatura especializada, em que pese a presença de vários estudos propondo modelos de previsão do consumo de eletricidade para os três segmentos residencial, industrial e comercial, apresenta majoritariamente modelos de tipo VAR e VCE. Este estudo atualiza a literatura no que concerne ao uso de modelos VAR e VCE e ao mesmo tempo os compara em termos preditivos com os modelos ARDL e estruturais a espaço de estados. Os resultados encontrados na análise do desempenho preditivo dos modelos mostraram que para o segmento residencial, o modelo com melhor capacidade preditivo foi o modelo estrutural, enquanto que para o segmento comercial foi o modelo VCE e, para o segmento industrial, foi o modelo ARDL. Previsões de 2014 a 2025 foram feitas com o intuito de informar ao mercado brasileiro a demanda de energia para cada segmento. Foram usadas bases de dados disponíveis e atualizadas provenientes das mesmas fontes usadas nos estudos da literatura.
Electricity is an input of widespread use in modern economies, penetrating in the most varied productive and consumption activities in society. However, the difficulties of large-scale storage make electricity very sensitive to supply conditions, to the point that supply problems quickly turns into blackouts. Among several devices implemented in the re-structuring of the Brazilian electricity sector (SEB) over the last 17 years, medium and long-term forecasting systems are used by public and private sector agents to reduce the uncertainties of the supply processes and expansion. ANEEL recommend in NT 292/2008-SER the use of three alternative multivariate methodologies in these prediction systems, namely: VAR and VCE models, autoregressive models with distributed lags (ARDL), and state space structural models. The specialized literature, despite the presence of several studies proposing models of prediction of the consumption of electricity for the three residential, industrial and commercial segments, mainly presents models of type VAR and VCE. This study updates the literature regarding the use of VAR and VCE models and at the same time compares them in predictive terms with the ARDL and structural state space models. The results found in the predictive model analysis showed that for the residential segment, the model with the best predictive capacity was the structural model, while for the commercial segment it was the VCE model and, for the industrial segment, it was the ARDL model. Forecasts from 2014 to 2025 were made with the intention of informing the Brazilian market the energy demand for each segment. Available and updated databases from the same sources used in literature studies were used.
41
Drélichová, Stanislava. "Studie řízení průběhu zakázky firmou." Master's thesis, Vysoké učení technické v Brně. Fakulta podnikatelská, 2007. http://www.nusl.cz/ntk/nusl-221414.
Full textAbstract:
This masterś thesis, titled as The Study of Order Process Control in Company Invensys Appliance Controls s.r.o., deals with optimization of order process from the first entering of new order till the delivery of finished goods to final customer. Further I provide the basic characteristics and analysis of Invensys Company. The project part is divided into two parts. The first one is focused on explaining of operation of Kanban system and itś possibility to improve the material flow in company. The second one is focused on long-term forecasts (demand predictions) provided by customers, which can help to improve the process of order as well. In conclusion, I appraise advantages and disadvantages of Kanban system and forecasts.
42
Steen, Englund Jessika. "Prediction of Energy Use of a Swedish Secondary School Building : Building Energy Simulation, Validation, Occupancy Behaviour and Potential Energy-Efficiency Measures." Licentiate thesis, Högskolan i Gävle, Energisystem och byggnadsteknik, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:hig:diva-33313.
Full textAbstract:
Residential and public buildings account for about 40% of the annual energy use in Europe. Many buildings are in urgent need of renovation, and reductions in energy demand in the built environment are of high importance in both Europe and Sweden. Building energy simulation (BES) tools are often used to predict building performance. However, it can be a challenge to create a reliable BES model that predicts the real building performance accurately. BES modelling is always associated with uncertainties, and modelling occupancy behaviour is a challenging task. This research presents a case study of a BES model of a school building from the 1960s in Gävle, Sweden, comprising an example of a validation strategy and a study of energy use and potential energy-efficiency measures (EEMs). The results show that collection of input data based on evidence, stepwise validation (for unoccupied and occupied cases), and the use of a backcasting method (which predicts varying occupancy behaviour and airing) is an appropriate strategy to create a reliable BES model of the studied school building. Several field measurements and data logging in the building management system were executed, in order to collect input data and for validation of the predicted results. Through the stepwise validation, the building’s technical and thermal performance was validated during an unoccupied period. The backcasting method demonstrates a strategy on how to predict the effect of the varying occupancy behaviour and airing activities in the school building, based on comparisons of BES model predictions and field measurement data. After applying the backcasting method to the model, it was validated during an occupied period. The annual predicted specific energy use was 73 kWh/m2 for heating of the studied building. The distribution of heat losses indicates that the best potential EEMs are changing to efficient windows, additional insulation of the external walls, improved envelope airtightness and new controls of the mechanical ventilation system.Byggnadssektorn står för ungefär 40 % av den årliga energianvändningen i Europa. Många byggnader är i stort behov av renovering och en minskning av energibehovet inom den byggda miljön är av stor vikt i både Europa och Sverige. För att undersöka byggnaders energianvändning används ofta simuleringsverktyg, men det kan vara utmanande att skapa pålitliga simuleringsmodeller som tillräckligt noggrant predikterar den verkliga byggnadens energianvändning. Simulering av byggnaders energianvändning är alltid förknippat med osäkerheter och att simulera människors beteendemönster är en stor utmaning. Den här forskningen innefattar en fallstudie med en simuleringsmodell av en skolbyggnad, byggd under 1960 talet och belägen i Gävle, inkluderat ett exempel på en valideringsstrategi och en studie av energianvändning och potentiella energieffektiviseringsåtgärder i byggnaden. Resultaten visar att insamling av indata baserade på evidens, stegvis validering (obemannad och bemannad) och användande av en backcasting-metod (vilket predikterar varierande brukarbeteende och vädring) är en lämplig strategi för att skapa en pålitlig energisimuleringsmodell för den studerade skolbyggnaden. Flertalet fältmätningar genomfördes och data loggades i systemet för fastighetsautomation, för att samla indata och för validering av de predikterade resultaten. Genom den stegvisa valideringen kunde byggnadens tekniska och termiska prestanda valideras för en obemannad period. Backcasting-metoden visar en strategi för hur man kan prediktera varierande brukarbeteende och vädringsaktiviteter i skolbyggnaden, baserat på jämförelser av modellens prediktioner och data från fältmätningar. När backcasting-metoden tillämpats i energisimuleringsmodellen, kunde modellen valideras för en bemannad period. Den årliga predikterade specifika energianvändningen för uppvärmningen är 73 kWh/m2. Fördelningen av värmeförluster i byggnaden indikerar att de bästa potentiella energieffektiviseringsåtgärderna är byte till fönster med bättre U-värde, tilläggsisolering av ytterväggarna, bättre lufttäthet i byggnadsskalet och ny styrning av det mekaniska ventilationssystemet.
43
Sassi, Kamal M. "Optimal scheduling, design, operation and control of reverse osmosis desalination : prediction of RO membrane performance under different design and operating conditions, synthesis of RO networks using MINLP optimization framework involving fouling, boron removal, variable seawater temperature and variable fresh water demand." Thesis, University of Bradford, 2012. http://hdl.handle.net/10454/5671.
Full textAbstract:
An accurate model for RO process has significant importance in the simulation and optimization proposes. A steady state model of RO process is developed based on solution diffusion theory to describe the permeation through membrane and thin film approach is used to describe the concentration polarization. The model is validated against the operation data reported in the literature. For the sake of clear understanding of the interaction of feed temperature and salinity on the design and operation of RO based desalination systems, simultaneous optimization of design and operation of RO network is investigated based on two-stage RO superstructure via MINLP approach. Different cases with several feed concentrations and seasonal variation of seawater temperature are presented. Also, the possibility of flexible scheduling in terms of the number of membrane modules required in operation in high and low temperature seasons is investigated A simultaneous modelling and optimization method for RO system including boron removal is then presented. A superstructure of the RO network is developed based on double pass RO network (two-stage seawater pass and one-stage brackish water pass). The MINLP problem based on the superstructure is used to find out an optimal RO network which will minimize the total annualized cost while fulfilling a given boron content limit. The effect of pH on boron rejection is investigated at deferent seawater temperatures. The optimal operation policy of RO system is then studied in this work considering variations in freshwater demand and with changing seawater temperature throughout the day. A storage tank is added to the RO layout to provide additional operational flexibility and to ensure the availability of freshwater at all times. Two optimization problems are solved incorporating two seawater temperature profiles, representing summer and winter seasons. The possibility of flexible scheduling of cleaning and maintenance of membrane modules is investigated. Then, the optimal design and operation of RO process is studied in the presence of membrane fouling and including several operational variations such as variable seawater temperature. The cleaning schedule of single stage RO process is formulated as MINLP problem using spiral wound modules. NNs based correlation has been developed based on the actual fouling data which can be used for estimating the permeability decline factors. The correlation based on actual data to predict the annual seawater temperature profile is also incorporated in the model. The proposed optimization procedure identified simultaneously the optimal maintenance schedule of RO network including its design parameters and operating policy. The steady state model of RO process is used to study the sensitivity of different operating and design parameters on the plant performance. A non-linear optimization problem is formulated to minimize specific energy consumption at fixed product flow rate and quality while optimizing the design and operating parameters. Then the MINLP formulation is used to find the optimal designs of RO layout for brackish water desalination. A variable fouling profile along the membrane stages is introduced to see how the network design and operation of the RO system are to be adjusted Finally, a preliminary control strategy for RO process is developed based on PID control algorithm and a first order transfer function (presented in the Appendix).
44
Neupane, Bijay. "Predictive Data Analytics for Energy Demand Flexibility." Doctoral thesis, Saechsische Landesbibliothek- Staats- und Universitaetsbibliothek Dresden, 2018. http://nbn-resolving.de/urn:nbn:de:bsz:14-qucosa-236309.
Full textAbstract:
The depleting fossil fuel and environmental concerns have created a revolutionary movement towards the installation and utilization of Renewable Energy Sources (RES) such as wind and solar energy. The RES entails challenges, both in regards to the physical integration into a grid system and regarding management of the expected demand. The flexibility in energy demand can facilitate the alignment of the supply and demand to achieve a dynamic Demand Response (DR). The flexibility is often not explicitly available or provided by a user and has to be analyzed and extracted automatically from historical consumption data. The predictive analytics of consumption data can reveal interesting patterns and periodicities that facilitate the effective extraction and representation of flexibility. The device-level analysis captures the atomic flexibilities in energy demand and provides the largest possible solution space to generate demand/supply schedules.
The presence of stochasticity and noise in the device-level consumption data and the unavailability of contextual information makes the analytics task challenging. Hence, it is essential to design predictive analytical techniques that work at an atomic data granularity and perform various analyses on the effectiveness of the proposed techniques. The Ph.D. study is sponsored by the TotalFlex Project (http://www.totalflex.dk/) and is part of the IT4BI-DC program with Aalborg University and TU Dresden as Home and Host University, respectively. The main objective of the TotalFlex project is to develop a cost-effective, market-based system that utilizes total flexibility in energy demand, and provide financial and environmental benefits to all involved parties. The flexibilities from various devices are modeled using a unified format called a flex-offer, which facilitates, e.g., aggregation and trading in the energy market. In this regards, this Ph.D. study focuses on the predictive analytics of the historical device operation behavior of consumers for an efficient and effective extraction of flexibilities in their energy demands.
First, the thesis performs a comprehensive survey of state-of-the-art work in the literature. It presents a critical review and analysis of various previously proposed approaches, algorithms, and methods in the field of user behavior analysis, forecasting, and flexibility analysis. Then, the thesis details the flexibility and flex-offer concepts and formally discusses the terminologies used throughout the thesis.
Second, the thesis contributes to a comprehensive analysis of energy consumption behavior at the device-level. The key motive of the analysis is to extract device operation patterns of users, the correlation between devices operations, and influence of external factors in device-level demands. A novel cost/benefit trade-off analysis of device flexibility is performed to categorize devices into various segments according to their flexibility potential. Moreover, device-specific data preprocessing steps are proposed to clean device-level raw data into a format suitable for flexibility analysis.
Third, the thesis presents various prediction models that are specifically tuned for device-level energy demand prediction. Further, it contributes to the feature engineering aspect of generating additional features from a demand consumption timeseries that effectively capture device operation preferences and patterns. The demand predictions utilize the carefully crafted features and other contextual information to improve the performance of the prediction models. Further, various demand prediction models are evaluated to determine the model, forecast horizon, and data granularity best suited for the device-level flexibility analysis. Furthermore, the effect of the forecast accuracy on flexibility-based DR is evaluated to identify an error level a market can absorb maintaining profitability.
Fourth, the thesis proposes a generalized process for automated generation and evaluation of flex-offers from the three types of household devices, namely Wet-devices, Electric Vehicles (EV), and Heat Pumps. The proposed process automatically predicts and estimates times and values of device-specific events representing flexibility in its operations. The predicted events are combined to generate flex-offers for the device future operations. Moreover, the actual flexibility potential of household devices is quantified for various contextual conditions and degree days.
Fifth, the thesis presents user-comfort oriented prescriptive techniques to prescribe flex-offers schedules. The proposed scheduler considers the trade-off between both social and financial aspects during scheduling of flex-offers, i.e., maximizing the financial benefits in a market and at the same time minimizing the loss of user comfort. Moreover, it also provides a distance-aware error measure that quantifies the actual performance of forecast models designed for flex-offers generation and scheduling.
Sixth, the thesis contributes to the comprehensive analysis of the financial viability of device-level flexibility for dynamic balancing of demand and supply. The thesis quantifies the financial benefits of flexibility and investigates the device type specific market that maximizes the potential of flexibility, both regarding DR and financial incentives. Henceforth, a financial analysis of each proposed technique, namely forecast model, flex-offer generation model, and flex-offer scheduling is performed. The key motive is to evaluate the usability of the proposed models in the device-level flexibility based DR scheme and their potential in generating a positive financial incentive to markets and customers.
Seven, the thesis presents a benchmark platform for device-level demand prediction. The platform provides the research community with a centralized repository of device-level datasets, forecast models, and functionalities that facilitate comparisons, evaluations, and validation of device-level forecast models.
The results of the thesis can contribute to the energy market in materializing the vision of utilizing consumption and production flexibility to obtain dynamic energy balance. The developed demand forecast and flex-offer generation models also contribute to the energy data analytics and data mining fields. The quantification of flexibility further contributes by demonstrating the feasibility and financial benefits of flexibility-based DR. The developed experimental platform provide researchers and practitioners with the resources required for device-level demand analytics and prediction.
45
Dyer, Ross. "Predicting residential demand: applying random forest to predict housing demand in Cape Town." Master's thesis, University of Cape Town, 2018. http://hdl.handle.net/11427/29602.
Full textAbstract:
The literature shows that Random Forest is a suitable technique to predict a target variable for a household with completely unseen characteristics. The models produced in this paper show that the characteristics of a household can be used to predict the Type of Dwelling, the Tenure and the Number of Bedrooms to varying degrees of accuracy. While none of the sets of models produced indicate a high degree of predictive accuracy relative to hurdle rates, the paper does demonstrate the value that the Random Forest technique offers in moving closer to an understanding of the complex nature of housing demand. A key finding is that the Census variables available for the models are not discriminatory enough to enable the high degree of accuracy expected from a predictive model.
46
Lin, Chuan-Heng, and 林泉亨. "MRT Demand Prediction through Social Media." Thesis, 2015. http://ndltd.ncl.edu.tw/handle/08325576544480906430.
Full textAbstract:
碩士國立臺灣大學
土木工程學研究所
103
With the technological improvements of mobile devices and the increasing number of social media posts, there are more and more data on human mobility based on which information could potentially be extracted. Current research related to social media are mostly focused on inter-person behaviors. Conversely, related topics on system level performances are rarely discussed. This thesis applies feature extraction methods on quantitative, textual, and image data to retrieve useful features from social media. In addition, a machine learning pipeline based on support vector machine, random forest and stochastic gradient boosting is constructed for a short-term transportation demand forecast. Furthermore, real-world datasets from Instagram together with the demand data of the Taipei Metro Rapid Transit system are demonstrated in this work. Validation results show that social media has the potential to enhance the forecasting accuracy.
47
LIANG, WAN YU, and 萬友良. "Prediction of Potential Demand on Mass Transportation." Thesis, 2012. http://ndltd.ncl.edu.tw/handle/44g673.
Full textAbstract:
碩士大葉大學
工業工程與科技管理學系
101
The object of this study is to set up logistics regression models to predict the probability that people shift to use mass-transportation by making use of major energy issues and the characteristics of mass-transportation as independent variables under the rise of gas price. First, the principal component analysis is employed to extract the major issues in energy. Secondly, canonical correlation analysis is applied to validate the relation between the major energy issues and the characteristics of mass-transportation. Finally, we surveyed the people around Tai-Chung area, and utilized logistic regression model to predict the probability that people shift to use mass-transportation. The results are followings：(1) People without driving license are more likely to use mass-transportation than the ones with it, while the gas price was risen up to 10%.（2）Males are more likely to use mass-transportation than females, while the gas price was risen up to 50%. (3) Singles are more likely to use mass-transportation than the married, while the gas price risen more than 100%. (4) Energy-saving and economical characteristics are the most important dependent variables which are the key words to increase the potential demand of mass-transportation, while convenience is not.
48
Cheng, Chih-Hsien, and 鄭志賢. "A Study on Domestic Water Demand Prediction." Thesis, 2008. http://ndltd.ncl.edu.tw/handle/73008774254015920073.
Full textAbstract:
碩士義守大學
資訊管理學系碩士班
96
Today, enterprises are forced to confront the challenges of global competition. They have had to grasp, and promptly make use of information ever since the Internet triggered more intensive competition in terms of time and space. Correctly anticipating client needs and quickly adapting to supply-demand trends are the foundation for enterprise''s sustainable development in the age of information explosion. In view of the nation’s self-awareness concerning raising economic, competitive advantages and the necessity of satisfying water resources demand, this research aims at constructing a prediction model of water demand and introducing: a water utilization pattern, a water consumption habit and an economic structure covering the next few years, in order to estimate water demand for water companies reference, and to benefit the nation, enterprises and people by managing water consumption effectively, resolving water shortage problems, and reducing the risk of domestic water use limitations or shortages. Although there are numerous models of water supply-demand, most researches have focused on supply-demand of annual water consumption. The Grey Prediction Model, of Grey''s theory, was used to predict the monthly water consumption in 12 branches of the Water Corporation, unlike the conventional prediction based on annual water consumption. The “uncertainty with little data” of Grey''s theory truly has more advantages, in terms of prediction, since the factors influencing water consumption were diversified and long-term historical statistics could not reflect current water consumption. The empirical results of the water consumption prediction model showed that according to average accuracy rate, the accuracy rate in using monthly water consumption data was above 95.7635% and the accuracy rate of five points rolling Grey Prediction Model reached up to 96.4274%, with more stable prediction results.
49
"A prediction model for short term electricity demand." Chinese University of Hong Kong, 1990. http://library.cuhk.edu.hk/record=b5886391.
Full textAbstract:
by Yung Kai-man.Thesis (M.B.A.)--Chinese University of Hong Kong, 1990.
Bibliography: leaves 90-92.
ABSTRACT --- p.ii
ACKNOWLEDGMENTS --- p.iii
TABLE OF CONTENTS --- p.iv
LIST OF ILLUSTRATIONS --- p.vi
LIST OF TABLES --- p.vii
Chapter
Chapter I. --- INTRODUCTION --- p.1
Background --- p.1
Methodology Review --- p.6
Chapter II. --- DATA BASE AND VARIABLES --- p.8
The Data Base --- p.8
The Dependent Variables --- p.9
The Independent Variables --- p.14
Chapter III. --- METHODOLOGY --- p.24
Regression Analysis --- p.24
Selection of the Predictors --- p.25
Regression Studies Using Moving Data --- p.29
Programming Aids --- p.32
Chapter IV. --- RESULTS AND DISCUSSIONS --- p.35
Validity of the Assumptions for the Regression Model --- p.35
Prediction Power of the Model --- p.37
Utility of the Prediction Model --- p.39
A Practical View of the Model Prediction --- p.47
Representation of the Predictors --- p.48
Chapter V. --- CONCLUSION AND RECOMMENDATIONS --- p.51
Evaluation of the Prediction Model --- p.51
Extension of the Project --- p.53
APPENDICES --- p.55
REFERENCES --- p.90
50
Chen, Shu-Pei, and 陳淑佩. "Prediction of train demand using advanced booking data." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/56120042217828403297.
Full textAbstract:
碩士國立成功大學
交通管理學系碩博士班
94
Booking curve is mostly used in Revenue management forecasting. Demand forecast is the core of revenue management. In literature review, improving 20% forecasting accuracy can increase 1% revenue. The objective of this research is to develop and test new models of train demand using advanced booking data. Train service offers several kinds of ticket sale including internet ticketing service, voice ticketing service, window ticketing service and instant ticketing sales. The research analyzes the way people book tickets in long trip. There are 73% tickets booked in advance but cancelled a lot. Many variables affect the pattern of booking curves including departure time, booking limit, day of week, holidays, and special days. Booking curves are different and the day of week affects the pattern of booking curves mostly. We use three kinds of data including unconstrained data, constrained data, and changed constrained data. In addition to the widely used models in revenue management including exponential smoothing model, mean of final bookings, simple linear regression, and pickup model, we develop several k-nearest neighbor models. According to the empirical analysis using Taiwan Railway Administration booking data, the proposed method provides promising results, e.g. the method improves the forecasting accuracy by 5% MAPE for the case of risk seeking.