Change search
Refine search result
1 - 19 of 19
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Rows per page
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sort
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
  • 1.
    Bengtsson, Tobias
    Blekinge Institute of Technology, Faculty of Engineering, Department of Industrial Economics.
    Comparison between proactive block replacement with no inventory and separate reactive replacement with inventory2017Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    To become a successful company today all costs, must be kept to a minimum. To make sure they are companies need to try new methods and policies to get closer to an optimum production. One of the things that need attention is the inventory of spare parts and the replacement of the same. The companies want their machines to have as many active hours as possible and therefore they need to replace components in them from time to time.

    This study will compare the current policy with a new one to see if it is economically justifiable to minimize the inventory of spare parts. This will be done by replacing two identical components in the same machine before they fail and production stops. We call it the proactive block replacement policy. To test this new policy a simulation based on historical data was made where the costs associated with the different replacements such as cost of component, cost o lost production and cost of the two types of replacements. All these costs were considered to see is the company could save money through this change.

    The result showed that the new policy was not suitable for this specific component because the cost of the component and the variation of lifetime on it was too high. Because of the big variation of lifetime, the time of the replacement had to be set after fewer hours of production which means more components will be used and the cost per active hour will be higher.

    This study is limited because only one specific component at a specific machine was studied and it is not possible to make any assumptions for other components from this study. This had to be done to get the most precise information from the company to get the best result.

    The conclusion of this study is that the company should keep their current replacement theory and not change into the new one. Though there might be possible ways of lowering the costs by only having one component in inventory instead of two.

  • 2.
    Engström, Rickard
    Blekinge Institute of Technology, Faculty of Engineering, Department of Mechanical Engineering.
    Optimering orderplock i ett mindre företag2017Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    The project was carried out at Mirro AB, a small company that manufactures sliding door solutions in its production in Hillerstorp, Småland. Within the company there are two brands, Mirro and Pelly System. The company has faced a challenge due to the fact that that there has been a big increase in volumes. This has led to consequences such as an increased number of picking errors for both brands as well as queues in Mirros area for compilation of orders. The purpose has been to suggest cost-effective solutions to these problems.

    One delimitation is that the complaints have been evaluated solely for the full calendar year of 2016. The study has only been limited to the picking stock for the brands Mirro and Pelly System. Therefore, the study has not taken into account any improvements needed in the production itself in addition to the proposed layout changes proposed to create better conditions for the flow of the stock. The study has been limited to the picking stock and the area for compilation of orders near by the picking stock.

     

    The two questions that were answered are:

     

    a)      How can picking errors from the stock at a small company be minimized?

    b)      In a small company with customer unique products, how can queuing in the area for compiling orders, be avoided by optimizing order picking in relation to the production flow?

    Case study regarding the company Mirro AB has been carried out with data collection through interviews and review of statistical material for complaints and storage transactions. This has since been used in various forms of systematic processing through SWOT analyses, brainstorming, Pareto diagrams and Ishikawa diagrams.

     

    Important results have been that a new, more efficient layout has been suggested for the warehouse that leads to more efficient order picking and order compilation. It is recommended to introduce barcode scanners with display and handheld computers as picking help and to reallocate personnel resources between the two brands to get a more even division of labour.

     

    The recommendations of the studies will help the company to correct its problems regarding the two issues without having to increase its resources. There will also be a better working situation due to the improved layout.

    The conclusion is that if the company implements the recommended improvement measures proposed, it will have a good possibility to correct the two issues. The result from the case study provided the expected results, it required more changes than I had expected.

     

    Key words: Order picking, warehouse, error picking, article 

  • 3.
    Gholami, Omid
    et al.
    Blekinge Institute of Technology, Faculty of Computing, Department of Computer Science and Engineering.
    Törnquist Krasemann, Johanna
    Blekinge Institute of Technology, Faculty of Computing, Department of Computer Science and Engineering.
    A Heuristic Approach to Solving the Train Traffic Re-Scheduling Problem in Real Time2018In: Algorithms, ISSN 1999-4893, E-ISSN 1999-4893, ISSN 1999-4893, Vol. 11, no 4, p. 1-18, article id 55Article in journal (Refereed)
    Abstract [en]

    Effectiveness in managing disturbances and disruptions in railway traffic networks, when they inevitably do occur, is a significant challenge, both from a practical and theoretical perspective. In this paper, we propose a heuristic approach for solving the real-time train traffic re-scheduling problem. This problem is here interpreted as a blocking job-shop scheduling problem, and a hybrid of the mixed graph and alternative graph is used for modelling the infrastructure and traffic dynamics on a mesoscopic level. A heuristic algorithm is developed and applied to resolve the conflicts by re-timing, re-ordering, and locally re-routing the trains. A part of the Southern Swedish railway network from Karlskrona centre to Malmö city is considered for an experimental performance assessment of the approach. The network consists of 290 block sections, and for a one-hour time horizon with around 80 active trains, the algorithm generates a solution in less than ten seconds. A benchmark with the corresponding mixed-integer program formulation, solved by commercial state-of-the-art solver Gurobi, is also conducted to assess the optimality of the generated solutions.

  • 4.
    Hallberg, Unni
    Blekinge Institute of Technology, Faculty of Engineering, Department of Industrial Economics.
    Effektivisering av materialpåfyllnad för indirekt material: En case study på företaget Alfa Laval Lund AB2015Independent thesis Advanced level (degree of Master (One Year)), 20 credits / 30 HE creditsStudent thesis
    Abstract [sv]

    Denna rapport tar upp undersökningen av effektivisering av materialpåfyllnad som utförts på företaget Alfa Laval Lund AB i Ronneby. Undersökningen har involverat förbättringsåtgärder för emballagematerialets flöde. Hur detta material kan hanteras med rätt material, i rätt tid och rätt kvantitet på ett tidseffektivt sätt utgör frågeställningen för denna undersökning. Då företaget idag använder sig av leanfilosofin utgör denna en stor del i det teoretiska kapitlet. Teorier om hur lean fungerar samt hur logistiken är kopplad till förbättring av materialflödet har ett teoretiskt kapitel där teorin förklarar hur dessa bör utformas. Även ett kapitel om hur metoderna observationer, intervjuer och brainstorming med flera används och varför finns också beskrivet. En förklaring om hur företaget arbetar idag och hur de involverade stationerna påverkar materialflödet tas också upp. Idag genomförs många kontrollrundor för att enbart kolla om produktionen har det emballagematerial som de behöver eller om något saknas. Dessa rundor sker flera gånger om dagen och resulterar inte i något värde för varken kunden eller företaget. Rundorna utgör en stor del i grunden till problemformuleringen, tillsammans med det andra problemet som handlar om kommunikation. Hur kan denna lösas då kontrollrundorna elimineras? Här har teorin om hur man bör signalera om behov genom dragande system istället för tryckande system som görs idag använts. Denna ständiga ovetskap om produktionsflödet materialbehov skapar stor stress för de anställda. Genom att spendera tid på att kontrollera samt kommunisera på ett sätt som tar upp mer tid än vad som behövs, flyter inte flödet på i ett dragande system.

    Tre olika åtgärdsförslag togs fram från metodgenomförandet. Dessa är följande:

     organisera lagret där emballagematerialet förvaras

     använda sig av ett flödeslager där man fyller på material som är uträknat att räcka för tre dagar

     en elektronisk kanbansignal som används för att signalera behov

    Slutsatsen för undersökningen är att dessa åtgärder effektiviserar materialpåfyllnaden genom att eliminera kontrollrundorna. Detta sparar tid och minskar stressen för de anställda genom att man nu får informationen till sig när produktionen har behov av material istället för att komma ihåg att kontrollera materialet med jämna mellanrum. Produktionen kan använda rätt materialet i rätt tid och har rätt kvantitet att tillgå. 

  • 5.
    Hantoft, Jonas
    Blekinge Institute of Technology, Faculty of Engineering, Department of Mechanical Engineering.
    Optimization of kitting process: A case study of Dynapac Compaction Equipment AB2015Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    A case study has been done at Dynapac Compaction Equipment AB in Karlskrona in order to improve the internal flow of the production. The “Supermarket Storage”, an adjoining storage that feed material to the lean production in the “Z-line” assembly line with the help of kitting, was chosen to be focused during the optimization of the internal flow. Also, due to the little academic research about kitting it was decided to focus the research on the kitting process and identify how to optimize it.

    The purpose of the research is to determine optimization methods of a kitting process and fill in the gap in the subject field about kitting optimization. Given the research time limit, the focus was only on the kitting process in the Supermarket Storage and no optimization could change the storage’s layout. This resulted in three research question that will be investigated in the thesis.

     Which common approaches exist when it comes to optimizing a kitting process?

     What is the result of each optimizing method in the time aspect?

     When should an optimization method be used, compared to the other methods that will be tested in this research?

    In order to solve these questions, was a needfinding process used in order to identify the kitting process current problems and the needs of the employees. With this, three optimization methods were identified and selected to be used to optimize the kitting process; optimization of routing, optimization of family grouping and optimization of an electronic system. The optimization of routing focused on the route that the kitters travel and the optimization of the family grouping focused on the article distribution in the Supermarket Storage; there each kitting operation’s articles should be stored in the same zone. Finally, the optimization of the electronic system, investigated the possibility to utilize a pick to scan system with the kitting process.

    Each optimization was implemented in different field experiment in order to identify how each optimization affected the kitting process. This resulted in that each optimization had improved the kitting process time efficiency and the electronic system had the biggest impact. Some other results were also observed during the experiments. The route optimization improved the learning curve of the kitting process and the family grouping optimization decreased the bottlenecks in kitting process. The electronic system optimization also implemented new benefits that resulted in a profit 2.5 times the cost of the system. Some of the benefits include removal of unneeded processes, quality control of the kitting process and statistics gathering that can be used to improve the process in the future.

    These results imply that all three optimization methods can be used in order to improve the time efficiency of a kitting process in a similar storage layout. The routing optimization should be used in a kitting operation with a high rotation of new kitters. The family grouping should be used in a kiting process with bottlenecks in the process and low organization of the article distribution. Ultimately, the electronic system optimization should be used in a kitting process that has unneeded processes and has the need of new tools that the electronic system can implement.

  • 6. Hjalmarsson, Oskar
    Miljökonsekvenser och kostnader för byte till elbilar inom Postnord i Vetlanda2016Independent thesis Basic level (professional degree), 10 credits / 15 HE creditsStudent thesis
  • 7.
    Karlsson, Louise
    Blekinge Institute of Technology, Faculty of Engineering, Department of Mechanical Engineering.
    A product-oriented Product Service System for tracing materials on autonomous construction sites: A product development for today’s and future construction sites2018Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    The global population is growing, and more people than before are moving to cities. This creates a need for increased building efficiency and possibility to work in remote environments. On today’s construction sites, there is a need to able to organize the site in a better way. In the future, autonomous vehicles will instead find it difficult to localize materials on a construction site. The autonomous vehicles can localize themselves with cameras and sensors, but they do not know how to localize the materials and items.

    This report is based on a project where Volvo Construction Equipment acted as a customer and the project was performed by students from Blekinge Institute of Technology and Stanford University. The prompt for this project was “From elephants to ants – from Earth to Mars” and would later be interpreted as finding a solution for the future that will be able to function without human’s intervention.

    From this project, this report was created. The following research questions for this report were:

    • How can workers locate building materials on today’s construction sites?

    • How will autonomous vehicles be able to locate material without human assistance in future construction sites?

    To solve these problems a design-process started, using an engineering design method. This method was chosen because of the type of problem. In engineering, the problem is identified to create a solution to the problem, comparing to when studying science, a question should be answered.

    The outcome from this report is a Product Service System (PSS) for a tracking system and a device for materials on today’s and future construction sites. When this solution was created no economic aspects were considered. Also, the focus of this report is the first steps of going from today’s construction sites to the future construction sites where autonomous vehicles will be used.

    The result from this research shows that the same problem of organizing a construction site is a pattern that can be seen in the majority of the sites that were visited during field works. Also, the workers today have little trust in the autonomous vehicles which is a result of lacking information and communication within companies. Furthermore, to be able to move to an autonomous future the mindset and attitude has to be changed. The collected data was analysed, and the outcome was a tracing system that will enable, both humans and machines, to localize materials on today’s and future construction sites. With this solution, today’s workers can track their materials wherever it is placed, without any need of changing the site. The autonomous vehicles will be able to use the tags to localize materials when there are no humans around.

  • 8.
    Lindahl, Emelie
    Blekinge Institute of Technology, Faculty of Engineering, Department of Mechanical Engineering. Carlsberg Sverige.
    Quality Investigation of Goods in the Beverage Industry: - A case study on continuous quality improvements in a warehouse2015Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Despite several developments in logistics that strive to decrease costs, tendencies can be found for increased logistical costs (Pewe, 2011, p. 17). Some factors causing increased system complexity are increased selection and faster market movements. The Warehouse Department at Carlsberg Sverige in Falkenberg has been chosen to be studied regarding continuous quality improvements. Carlsberg Sverige implemented an automated inventory system and automated order-picking system in 2012-2013, which has had about 40% increase of stock keeping units (SKUs) since the investments. Since the implementation of the automation Carlsberg Sverige has encountered challenges with other quality of goods. The system has become more sensitive and it has become more important to eliminate deviation.The study aimed to develop a way for long-term reduction or elimination of quality errors that result in negative effects to the system. A specific area investigated was regarding both effect of the organization in a holistic perspective and on a level of detail of the analyzed area. The study did not include quality of beverages, but only external quality problems of goods, such as faulty stacking of packages or plastic wrapping hanging outside of the goods.An exploratory study was conducted with predominantly quantitative data collection methods. Initially a current state mapping was made, a flow chart was created through interviews and observations of employees. Critical activities / situations were identified in the flow chart and three areas were determined for further investigation regarding quality errors. Observations were carried out where all occurred abnormalities were registered. The results were analyzed and a focus area determined for continued deeper examination. A final model was created with the influence of theories and collected data from the case study.The examined focus area was part of the fully automated warehouse. Adaptation to the new automation was still under investigation and a need for more clarity in the continuous quality improvements . From the observations, it was revealed that in approximately 70% of the observing time in the focus area, there was an error in the area affected system. Registered errors were categorized into ten groups, seven groups were included in the research scope. Out of the seven categories there were two significance regarding amount of repetitions of registered errors; unreadable label was measured 30 times, and plastic wrapping outside goods was discovered 12 times during 7 hours of measurement. Significance regarding average duration was discovered for; system errors of labels and wood detected by sensors on conveyors.Detected errors with significance were further analyzed with fishbone diagrams to find root-causes. Several common causes were discovered in the fishbone diagrams, for instance regarding inadequate methods and measurements. The discoveries functioned as basis for the final model that consists of: Key Performance Indexes (KPI's), a flow chart of affecting processes, daily whiteboard meetings, improved communication channels and a process for implementation of routines. The model developed was to fit this case study, which represents; a complex system with lots of shift work, both manual and automated processes and handling of beverage goods. Despite this, the model is considered to function as a basis to work on continuous quality improvements for organizations outside the mentioned scope, but with modifications of the model.

  • 9.
    Liu, Qiyang
    et al.
    Blekinge Institute of Technology, Faculty of Engineering, Department of Strategic Sustainable Development.
    Han, Yini
    Blekinge Institute of Technology, Faculty of Engineering, Department of Strategic Sustainable Development.
    Liddawi, Shafiq
    Blekinge Institute of Technology, Faculty of Engineering, Department of Strategic Sustainable Development.
    Key Factors of Public Attitude towards Sustainable Transport Policies: A Case Study in Four Cities in Sweden2015Independent thesis Advanced level (degree of Master (One Year)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Urban transport systems are facing seemingly irreconcilable problems.Sustainable transport policies are necessary to address this sustainability challenge. However,their effectiveness highly depends on the public attitude towards them. A method combiningboth qualitative and quantitative research was used to redefine a sustainable transport policybased on sustainability principles, and find out the key factors of public attitude towardssustainable transport policies. Furthermore, the interrelations between these factors arerevealed by using structural equation modelling. 1685 questionnaires were sent to fourrepresentative cities in Sweden: Stockholm, Göteborg, Uppsala and Karlskrona. By using thisfirst hand data, we discovered some differences between Karlskrona and these other threecities. The negative attitudes towards reducing car use and promoting public transport inKarlskrona is opposite to the positive attitudes in other cities. The acceptance of taxes, speedlimitation and parking regulation is notably different as well. The results also indicate that citycharacteristics influence public attitudes towards sustainable transport policies more thancitizens’ characteristics. The functionality of a city is the most significant factor. Moreover,the results show a high dependence on individual car use. This suggests that planners shouldnot use the experience gained from other cities without investigating actual local conditions.

  • 10.
    Paulauskas, Vytautas
    et al.
    Klaipeda Shipping Research Centre, LTU.
    Henesey, Lawrence
    Blekinge Institute of Technology, Faculty of Computing, Department of Computer Science and Engineering.
    Paulauskas, Donatas
    Klaipeda University, LTU.
    Ronkaitytė, Leva
    Klaipeda University, LTU.
    Gerlitz, Laima
    Wismar University of Applied Sciences, DEU.
    Jankowski, S.
    Akademia Morska w Szczecinie, POL.
    Canepa, M.
    World Maritime University, SWE.
    LNG bunkering stations location optimization on basis graph theory2018In: Transport Means - Proceedings of the International Conference, Kaunas University of Technology , 2018, p. 660-664Conference paper (Refereed)
    Abstract [en]

    As an alternative to traditional fuel and energy source LNG (Liquefied Natural Gas) has many advantages, such as lower emissions while providing a means of energy for trucks, trains and ships. In focusing on the maritime transport sector the reasons for using LNG make a convincing business case, but lead to many discussions on LNG investments. The key issue has been is: “should investment be implemented first on LNG bunkering stations and then wait for the market to build ships to use the facilities or should investment wait until there is a demand?” Obviously, this creates a “chicken-and-egg” situation on when and where to invest for LNG use to take place. The initial experiences in using LNG in maritime and road transport suggest that the transport firms often take the risk themselves by not only investing into the transport units (ships, trucks) themselves but also invest into the infrastructure as well, e.g., developing LNG bunkering facilities. At the same time with these large initial investments for developing LNG bunkering networks there are more and more requests for identifying optimal solutions, often are based on real LNG fuel demand in ports and on the roads. This paper is oriented on the study for optimal bunkering network creation, which is argued to help with improved efficiency in the supply of LNG fuel to transport users. In addition, optimal investments for LNG bunkering networks can be realized. © 2018 Kaunas University of Technology. All rights reserved.

  • 11.
    Paulauskas, Vytautas
    et al.
    Klaipeda University, LTU.
    Paulauskas, Donatas
    Klaipeda University, LTU.
    Placiene, Birutė
    Klaipeda University, LTU.
    Barzdziukas, Raimondas
    Klaipeda University, LTU.
    Maksimavicius, Ričardas
    Klaipeda University, LTU.
    Ronkaityte, I.
    Klaipeda University, LTU.
    Gerlitz, Laima
    Wismar University of Applied Sciences, DEU.
    Madjidian, J.
    World Maritime University, SWE.
    Jankowski, S.
    Akademia Morska w Szczecinie, POL.
    Henesey, Lawrence
    Blekinge Institute of Technology, Faculty of Computing, Department of Computer Science and Engineering.
    Optimization modelling of LNG supply chains for development: Case study of Lithuania and Latvia2017In: Transport Means - Proceedings of the International Conference, Kaunas University of Technology , 2017, p. 762-765Conference paper (Refereed)
    Abstract [en]

    The increasing demand for Liquid Natural Gas (LNG) is causing many challenges for users and suppliers worldwide. Though there is strong interest in using LNG, the research published in this paper indicates there are challenges in developing adequate delivery and distribution chains within the supply chain. Ideally, LNG distribution chains should be created on the basis of user demands and need. In this paper we have articulated an optimisation model that considers the various potential users and their characteristics in order to identify if possibilities and prospects exist in developing an adequate LNG supply chain. The case study of Lithuania and Latvia serves as a model from which we are able to use our tool to help identify the factors for success in creating such LNG supply chains. © 2017 Kaunas University of Technology. All rights reserved.

  • 12.
    Schulte, Jesko
    et al.
    Blekinge Institute of Technology, Faculty of Engineering, Department of Strategic Sustainable Development.
    Ny, Henrik
    Blekinge Institute of Technology, Faculty of Engineering, Department of Strategic Sustainable Development.
    Electric road systems: Strategic stepping stone on the way towards sustainable freight transport?2018In: Sustainability, ISSN 2071-1050, E-ISSN 2071-1050, Vol. 10, no 4, article id 1148Article in journal (Refereed)
    Abstract [en]

    Electrification of the transport sector has been pointed out as a key factor for tackling some of today's main challenges, such as global warming, air pollution, and eco-system degradation. While numerous studies have investigated the potential of electrifying passenger transport, less focus has been on how road freight transport could be powered in a sustainable future. This study looks at Electric Road Systems (ERS) in comparison to the current diesel system. The Framework for Strategic Sustainable Development was used to assess whether ERS could be a stepping stone on the way towards sustainability. Strategic life-cycle assessment was applied, scanning each life-cycle phase for violations against basic sustainability principles. Resulting sustainability "hot spots" were quantified with traditional life-cycle assessment. The results show that, if powered by renewable energy, ERS have a potential to decrease the environmental impact of freight transport considerably. Environmental payback times of less than five years are achievable if freight traffic volumes are sufficiently high. However, some severe violations against sustainability principles were identified. Still, ERS could prove to be a valuable part of the solution, as they drastically decrease the need for large batteries with high cost and sustainability impact, thereby catalyzing electrification and the transition towards sustainable freight transport. © 2018 by the authors.

  • 13.
    Sigakova, Ksenia
    et al.
    Blekinge Institute of Technology, Faculty of Computing, Department of Computer Science and Engineering.
    Mbiydzenyuy, Gideon
    Blekinge Institute of Technology, Faculty of Computing, Department of Computer Science and Engineering.
    Holmgren, Johan
    Blekinge Institute of Technology, Faculty of Computing, Department of Computer Science and Engineering.
    Impacts of traffic conditions on the performance of road freight transport2015In: 2015 IEEE 18TH INTERNATIONAL CONFERENCE ON INTELLIGENT TRANSPORTATION SYSTEMS, 2015, p. 2947-2952Conference paper (Refereed)
    Abstract [en]

    The efficiency of road transport is typically influenced by factors such as, weather, choice of road, and time of day, and day of the week. Knowledge about interactions between different traffic-and transport related factors and their influence on the execution of transport is important in transport planning. The purpose of this paper is to study the impact of different factors on the performance of road transport. We aim to contribute to improved transport planning by analysing traffic and transport data obtained from different sources in order to support data driven decision making. Through a review of existing literature and discussions with a Swedish road transport operator, we identified factors that could be relevant to consider when planning a transport, e.g., related to weather, location of roads where the transport will take place, and planned time of the transport. As a result of variation in size, type and volume of the data representing these factors, suitable machine learning algorithms were selected, such as Decision Stump, M5 model tree, M5 regression tree, RepTree, M5 rules, and linear regression in order to study the data. Our experimental results illustrate the complexity associated to the performance of road transport systems mainly because of the dependency between the choices of influencing factors and geographic location of the road segment.

  • 14.
    Sun, Bin
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Automated Traffic Time Series Prediction2018Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    Intelligent transportation systems (ITS) are becoming more and more effective. Robust and accurate short-term traffic prediction plays a key role in modern ITS and demands continuous improvement. Benefiting from better data collection and storage strategies, a huge amount of traffic data is archived which can be used for this purpose especially by using machine learning.

    For the data preprocessing stage, despite the amount of data available, missing data records and their messy labels are two problems that prevent many prediction algorithms in ITS from working effectively and smoothly. For the prediction stage, though there are many prediction algorithms, higher accuracy and more automated procedures are needed.

    Considering both preprocessing and prediction studies, one widely used algorithm is k-nearest neighbours (kNN) which has shown high accuracy and efficiency. However, the general kNN is designed for matrix instead of time series which lacks the use of time series characteristics. Choosing the right parameter values for kNN is problematic due to dynamic traffic characteristics. This thesis analyses kNN based algorithms and improves the prediction accuracy with better parameter handling using time series characteristics.

    Specifically, for the data preprocessing stage, this work introduces gap-sensitive windowed kNN (GSW-kNN) imputation. Besides, a Mahalanobis distance-based algorithm is improved to support correcting and complementing label information. Later, several automated and dynamic procedures are proposed and different strategies for making use of data and parameters are also compared.

    Two real-world datasets are used to conduct experiments in different papers. The results show that GSW-kNN imputation is 34% on average more accurate than benchmarking methods, and it is still robust even if the missing ratio increases to 90%. The Mahalanobis distance-based models efficiently correct and complement label information which is then used to fairly compare performance of algorithms. The proposed dynamic procedure (DP) performs better than manually adjusted kNN and other benchmarking methods in terms of accuracy on average. What is better, weighted parameter tuples (WPT) gives more accurate results than any human tuned parameters which cannot be achieved manually in practice. The experiments indicate that the relations among parameters are compound and the flow-aware strategy performs better than the time-aware one. Thus, it is suggested to consider all parameter strategies simultaneously as ensemble strategies especially by including window in flow-aware strategies.

    In summary, this thesis improves the accuracy and automation level of short-term traffic prediction with proposed high-speed algorithms.

  • 15.
    Sun, Bin
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Toward Automatic Data-Driven Traffic Time Series Prediction2017In: 5th Swedish Workshop on Data Science, 2017, Vol. 12, article id 12Conference paper (Refereed)
    Abstract [en]

    Short-term traffic prediction on freeways has been an active research subject in the past several decades. Various algorithms covering a broad range of topics regarding performance, data requirements and efficiency have been proposed. However, the implementation of machine learning based algorithms in traffic management centres is still limited. Two main reasons for this situation are, the data is messy or missing, and the parameter tuning requires experienced engineers.

    The main objective of this thesis was to develop a procedure that can improve the performance and automation level of short-term traffic prediction.

    Missing data is a problem that prevents many prediction algorithms in ITS from working effectively. Much work has been done to impute those missing data. Among different imputation methods, k-nearest neighbours (kNN) has shown excellent accuracy and efficiency. However, the general kNN is designed for matrix instead of time series so it lacks the usage of time series characteristics such as windows and weights that are gap-sensitive. We introduce gap-sensitive windowed kNN (GSW-kNN) imputation for time series. The results show that GSW-kNN is 34% more accurate than benchmarking methods, and it is still robust even if the missing ratio increases to 90%.

    Lacking accurate accident information (labels) is another problem that prevents huge amount of traffic data to be fully used. We improve a Mahalanobis distance based algorithm to be able to handle differential data to estimate flow fluctuations and detect accidents and use it to support correcting and complementing accident information. The outlier detection algorithm provides accurate suggestions for accident occurring time, duration and direction. We also develop a system with interactive user interface to realize this procedure. There are three contributions for data handling. Firstly, we propose to use multi-metric traffic data instead of single metric for traffic outlier detection. Secondly, we present a practical method to organise traffic data and to evaluate the organisation for Mahalanobis distance. Thirdly, we describe a general method to modify Mahalanobis distance algorithms to be updatable.

    For automatic parameter tuning, the experiments show that the flow-aware strategy performs better than the time-aware one. Thus, we use all parameter strategies simultaneously as ensemble strategies especially by including window in flow-aware strategies.

    Based on the above studies, we have developed online-orientated and offline-orientated algorithms for real-time traffic forecasting. The online automatic tuned version is performing near the optimal manual tuned performance. The offline version gives the performance that cannot be achieved using the manual tuning. It is also 3.05% better than XGB and 11.7% better than traditional SARIMA.

  • 16.
    Sun, Bin
    et al.
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Cheng, Wei
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Goswami, Prashant
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Bai, Guohua
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Short-Term Traffic Forecasting Using Self-Adjusting k-Nearest Neighbours2018In: IET Intelligent Transport Systems, ISSN 1751-956X, E-ISSN 1751-9578, Vol. 12, no 1, p. 41-48Article in journal (Refereed)
    Abstract [en]

    Short-term traffic forecasting is becoming more important in intelligent transportation systems. The k-nearest neighbours (kNN) method is widely used for short-term traffic forecasting.However, kNN parameters self-adjustment has been a problem due to dynamic traffic characteristics. This paper proposes a fully automatic dynamic procedure kNN (DP-kNN) that makes the kNN parameters self-adjustable and robust without predefined models or training. We used realworld data with more than one-year traffic records to conduct experiments. The results show that DP-kNN can perform better than manually adjusted kNN and other benchmarking methods with regards to accuracy on average. This study also discusses the difference between holiday and workday traffic prediction as well as the usage of neighbour distance measurement.

  • 17.
    Sun, Bin
    et al.
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Liyao, Ma
    University of Jinan, CHI.
    Wei, Cheng
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Wei, Wen
    Blekinge Institute of Technology, Faculty of Computing, Department of Technology and Aesthetics.
    Prashant, Goswami
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Guohua, Bai
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    An Improved k-Nearest Neighbours Method for Traffic Time Series Imputation2017Conference paper (Refereed)
    Abstract [en]

    Intelligent transportation systems (ITS) are becoming more and more effective, benefiting from big data. Despite this, missing data is a problem that prevents many prediction algorithms in ITS from working effectively. Much work has been done to impute those missing data. Among different imputation methods, k-nearest neighbours (kNN) has shown excellent accuracy and efficiency. However, the general kNN is designed for matrix instead of time series so it lacks the usage of time series characteristics such as windows and weights that are gap-sensitive. This work introduces gap-sensitive windowed kNN (GSW-kNN) imputation for time series. The results show that GSW-kNN is 34% more accurate than benchmarking methods, and it is still robust even if the missing ratio increases to 90%.

  • 18.
    Sun, Bin
    et al.
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Wei, Cheng
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Prashant, Goswami
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Guohua, Bai
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    An Overview of Parameter and Data Strategies for K-Nearest Neighbours Based Short-Term Traffic Prediction2017In: ACM International Conference Proceeding Series Volume Part F133326, Association for Computing Machinery (ACM), 2017, p. 68-74Conference paper (Refereed)
    Abstract [en]

    Modern intelligent transportation systems (ITS) requires reliable and accurate short-term traffic prediction. One widely used method to predict traffic is k-nearest neighbours (kNN). Though many studies have tried to improve kNN with parameter strategies and data strategies, there is no comprehensive analysis of those strategies. This paper aims to analyse kNN strategies and guide future work to select the right strategy to improve prediction accuracy. Firstly, we examine the relations among three kNN parameters, which are: number of nearest neighbours (k), search step length (d) and window size (v). We also analysed predict step ahead (m) which is not a parameter but a user requirement and configuration. The analyses indicate that the relations among parameters are compound especially when traffic flow states are considered. The results show that strategy of using v leads to outstanding accuracy improvement. Later, we compare different data strategies such as flow-aware and time-aware ones together with ensemble strategies. The experiments show that the flowaware strategy performs better than the time-aware one. Thus, we suggest considering all parameter strategies simultaneously as ensemble strategies especially by including v in flow-aware strategies.

  • 19.
    Sun, Bin
    et al.
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Wei, Cheng
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Prashant, Goswami
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Guohua, Bai
    Blekinge Institute of Technology, Faculty of Computing, Department of Creative Technologies.
    Flow-Aware WPT k-Nearest Neighbours Regression for Short-Term Traffic Prediction2017In: Proceedings - IEEE Symposium on Computers and Communications, Institute of Electrical and Electronics Engineers (IEEE), 2017, Vol. 07, p. 48-53, article id 8024503Conference paper (Refereed)
    Abstract [en]

    Robust and accurate traffic prediction is critical in modern intelligent transportation systems (ITS). One widely used method for short-term traffic prediction is k-nearest neighbours (kNN). However, choosing the right parameter values for kNN is problematic. Although many studies have investigated this problem, they did not consider all parameters of kNN at the same time. This paper aims to improve kNN prediction accuracy by tuning all parameters simultaneously concerning dynamic traffic characteristics. We propose weighted parameter tuples (WPT) to calculate weighted average dynamically according to flow rate. Comprehensive experiments are conducted on one-year real-world data. The results show that flow-aware WPT kNN performs better than manually tuned kNN as well as benchmark methods such as extreme gradient boosting (XGB) and seasonal autoregressive integrated moving average (SARIMA). Thus, it is recommended to use dynamic parameters regarding traffic flow and to consider all parameters at the same time.

1 - 19 of 19
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf