published_or_final_version / Transport Studies / Master / Master of Arts

published_or_final_version / abstract / Transport Policy and Planning / Master / Master of Arts in Transport Policy and Planning

In this dissertation, we aim to investigate some delivery planning problems under different aspects of uncertain conditions in order to present some insights on how to develop more efficient and effective logistics solutions in the challenging, dynamic and competitive industry of express delivery nowadays. / Lastly, we introduce a travel time estimation method which can be considered as an upstream extension on the first delivery planning problem or any routing problems that require travel time in a road network as an input parameter. We realize that different trips may traverse on some links that are transited by other trips as well; and therefore linear and quadratic programming models can be developed to infer the travel time of any road segments which are traversed by more than one trip from the consideration of the common road segments of various trips. Bus routes information is proposed as the data source because of their coverage on the road network and their data available. Robustness of the proposed models is evaluated according to the deviation between the resultant inferences and the means of the segment travel times. It is found that the basic models suffer from a fundamental problem of underspecification issue if the number of observed trips (i.e. constraints) is fewer than the number of road segments (i.e. variables) to be estimated. Therefore, two additional types of constraints are introduced to address these issues. / The focus of the second delivery planning problem addressed in this dissertation is switched to the randomness of the demands at the delivery locations. We consider a single-depot problem in which a large volume of packages have to be delivered in a dense and small area with a lot of high-rise buildings within an extremely tight delivery commitment time window. At each vertex, the package volume (i.e. demand) is a highly stochastic variable and its service time is also dependent on the demand. This version of problem is typically faced by express service providers which offer premium express services to customers that are protected by the money back guarantee (MBG) condition. We name this version of problem as a time-constrained vehicle routing problem with stochastic demands and service times (SVRP-D). Since both the demand and service time of a vertex are stochastic, the vehicle capacity and the commitment-time (i.e. deadline) constraints may not be satisfied after the demands are realized, and recourse action is thus necessary to "rescue" the "infeasible" delivery plan through additional effort and cost. Three typical recourse actions are independently considered: (1) restocking; (2) outsourcing; and (3) reassignment. A two-stage stochastic integer programming is proposed such that the first-stage problem is to determine a set of a priori routes which minimize the vehicle travel cost and the cost for any known recourse actions. Dependent on a first-stage solution, the second-stage problem is to decide the optimal recourse plan with least cost according to the pre-determined recourse action. / Two delivery planning problems and one travel time estimation model under various aspects of randomness are addressed in this dissertation. The first delivery planning problem considers an extension on a well-known problem---the vehicle routing problem with time windows (VRPTW), where the travel time required between each delivery location is a stochastic variable, instead of a fixed value. We name this version of the VRPTW as a vehicle routing problem with time windows and stochastic travel times (VRPTWST). A two-stage stochastic integer programming with recourse model is developed and its objective is to minimize the vehicle travel cost together with the expected loss due to the violation of the delivery time windows. An exact branch-and-cut (B&C) algorithm is proposed. This algorithm is effective for small-size problems, e.g. instances with only 8 vertices. For larger-size problems, some modifications on the B&C algorithm are suggested to improve the solution time with only a small deviation from optimality. / Wong Chi Fat. / "May 2007." / Adviser: Janny Leung. / Source: Dissertation Abstracts International, Volume: 69-01, Section: B, page: 0596. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2007. / Includes bibliographical references (p. 256-270). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract in English and Chinese. / School code: 1307.

Thesis (M.B.A.)--University of Hong Kong, 1997. / Includes bibliographical references (leaf 68).

Thesis (M.A.)--University of Hong Kong, 1994. / Includes bibliographical references. Also available in print.

Submitted by Nadir Basilio (nadirsb@uninove.br) on 2016-07-01T14:42:02Z No. of bitstreams: 1 Julio Cezar De Oliveira.pdf: 1235519 bytes, checksum: 7e63ce47300a84dec3bc01d87650f12f (MD5) / Made available in DSpace on 2016-07-01T14:42:02Z (GMT). No. of bitstreams: 1 Julio Cezar De Oliveira.pdf: 1235519 bytes, checksum: 7e63ce47300a84dec3bc01d87650f12f (MD5) Previous issue date: 2016-02-29 / Outsourcing of logistics activities in the Brazil has been growing, allowing companies to direct efforts towards to their core activities and to obtain strategic gains due to the know-how of specialist companies and the possibility of reducing costs with the assets of the contracted company. Thus, tools that aid in the proper selection of suppliers can be important for the company to reach its goals. In this context, the purpose of this study is to apply the Analytic Hierarchy Process (AHP) in the evaluation process and selection of a logistics services providers (LSP) for the courier delivery services company, analyzing qualitative and quantitative criteria that express the real needs of the parties, whose aim is to improve the indicators in terms of quality and customer satisfaction. The results indicated that the information technology, responsiveness and quality of management are the most relevant criteria for the contractor because of the need of preserving the enterprise reliability front of your customers and to remain competitive in the courier delivery services area. The results also showed that the criteria and method used could contribute for classification of LSP suitable for the service in terms of deadline, demand and traceability required for operational success. A sensitivity analysis was also carried out to assess the robustness of the decision. Finally, the proposed model proved to be a good alternative to support the decision on the selection of logistics providers for express charge, since the analysis of qualitative and quantitative criteria adhering to organizational practice, can lead to choosing companies that can meet the quality requirements of the sector. / A terceirização das atividades logísticas no Brasil vem crescendo, permitindo as empresas a direcionar os esforços no sentido de suas atividades principais e obter ganhos estratégicos devido ao know-how de empresas especializadas e a possibilidade de reduzir os custos com os ativos da empresa contratada. Desta forma, ferramentas que auxiliem na seleção adequada de fornecedores, pode ser importante para que a empresa atinja seus objetivos. Nesse contexto, o objetivo deste estudo é investigar o uso do método de decisão multicritério Analytic Hierarchy Process (AHP) para a avaliação e seleção de provedores de serviços logísticos (PSLs) para empresas do segmento de cargas expressas analisando critérios qualitativos e quantitativos que expressam as reais necessidades das partes envolvidas, cujo intuito é melhorar os indicadores em termo de qualidade, prazo e satisfação dos clientes. Os resultados demonstraram que a tecnologia da informação, responsividade e a gestão da qualidade são os critérios mais relevantes para a contratante em razão da necessidade da preservação da confiabilidade da empresa frente a seus clientes e para manter-se competitiva no segmento de cargas expressas. Constatou-se que os critérios utilizados e o método contribuíram para a classificação de PSLs adequados para o atendimento em termos de prazo, demanda e rastreabilidade exigidos para o sucesso operacional. Uma análise de sensibilidade também foi desenvolvida para avaliar a robustez da decisão. O modelo proposto mostrou ser uma boa alternativa para apoio a decisão na seleção de fornecedores logísticos para carga expressa, uma vez que a análise de critérios qualitativos e quantitativos aderentes à prática organizacional, pode levar a escolha de empresas que possam atender as exigências de qualidade do setor.