Horizontal collaboration among carriers in dynamic truckload transportation

動態整車協同運輸優化運作方法研究

Student thesis: Doctoral Thesis

View graph of relations

Author(s)

  • Shujun LIU

Detail(s)

Awarding Institution
Supervisors/Advisors
Award date2 Oct 2013

Abstract

In the contemporary transportation industry, freight carriers often need to provide immediate transportation services so as to be able to cope with their customers' time constraints. In the truckload transportation market, each carrier has to manage its own trucks dynamically to fulfil requests fed to it continuously, in accordance with tight schedules. Meanwhile, due to demand uctuations, the carriers are usually bombarded with excess/shortage complaints. The problem is particularly exacerbated when, as is often the case, there is little information on future requests and no additional resources are available. In such a situation, prospects of cost reduction through individual order optimization become severely limited. In a dynamic truckload transportation environment, each carrier would like to instantly extend its own resource portfolios in order to cope with resource shortages. This is particularly feasible, if external, excess resources owned by other carriers are available at that particular point in time. In this thesis, the horizontal resource collaboration approach is examined to realize further transportation cost reductions by offering each participant carrier an opportunity to gain synergy profits. The approach is developed from an examination of practical situations and integrates game theory and mechanism design methods with optimization techniques. The collaborative approach is capable of improving resource utilization on the art of all participating carriers by balancing the required resources with the available. The method is found to be capable of helping participating carriers handle the dilemmas associated with instant resource shortages without losing individual autonomy. The main research issues here involved the identification and utilization of synergies between the resource instant utilization statuses of all participant carriers in a dynamic transportation environment wile ensuring a fair and efficient allocation of synergy benefits among them. The first part of the thesis addresses the commonly faced dynamic truckload transportation problem with capacity restriction (DTT). A noteworthy aspect of this problem concerns truck resource shortage. If no other trucks are available, requests beyond the carrier's own transportation capacity have to be rejected when a resource instant shortage occurs. A mathematical programming formula is introduced for the on-line version of the DTT problem. A column generation algorithm, embedded with a label-correcting algorithm, has been developed to solve the associated NP-hard problem. This formula is then used as the basis for more complicated resource-collaboration applications, and the solutions utilized, as benchmarks for subsequent analyses and evaluations. The second part of the thesis proposes two resource sharing strategies among a dominant carrier and its affiliates. Because the dominant carrier wants to maintain market position, rather than subcontract transportation jobs, it usually prefers to hire external trucks from affiliates. In practice, external trucks should be hired by the dominant carrier, only when resource instant shortage occurs. This makes the new, proposed dynamic vehicle routing problem different from the more common ones which which tend to reject excessive requests. A mathematical programming formulation constructed by a master problem and a sub-problem is proposed to describe the new problem. The column generation algorithm embedded with a label-correcting algorithm helps carriers cope with new constraints as they arise. Two resource collaboration strategies{resource collaboration on a daily basis and resource collaboration on route basis{are drawn following a study of several practical situations. Finally, the effectiveness of the strategies is evaluated. The last part of the thesis discusses resource sharing among several nondominant carriers within a dynamic truckload transportation environment. With individual autonomy, each carrier decides whether to participate in resource sharing before collaborative transportation starts. Resource sharing operates in two sequential procedures: resource sharing and profit allocation. A new collaborative and dynamic transportation problem, with due consideration of nondominant collaborative relationships among the participating carriers is defined. After the implementation of collaborative transportation, the solution method of nucleolus, using a constraint generation algorithm is applied to allocate synergistic profits.

    Research areas

  • Trucking, Management