Cost Allocation in Collaborative Transportation Defended on Friday, 12 November 2021
Significant cost reductions, both monetary and environmentally, can be realised by collaborative efforts in logistics. Specifically, we study centralised horizontal collaborations between logistic service providers in the field of transportation. In this particular setting, cost reductions are achieved by pooling requests, combining delivery routes and collectively serving new customers.
First, we consider a centralised horizontal collaboration between logistic service providers that is governed by a third party. Here, we assume that the collaborative operations are planned by this third party. We focus on how to determine a stable allocation of the overall cost or profit among the logistic service providers. That is, no subset of logistic service providers is better off by leaving the collaboration. For two different collaborative settings, we propose algorithms to determine allocations which ensure a stable collaboration among the logistic service providers, or to show that no such allocation exists. Moreover, we demonstrate the efficiency of these algorithms.
Second, we focus on a centralised horizontal collaboration that is not governed by a third party. That is, the logistic service providers need to plan their joint operations together. We propose a framework, based on cooperative game theory, in which the algorithms of each logistic service provider can be used to determine a planning for their joint operations. In doing so, any logistic service provider or third party with a good algorithm can add value to the collaboration, even without being a logistic service provider, i.e., a consultant. We study this setting both analytically and numerically for collaborations in transportation, and provide insights into the value of the quality of an algorithm in collaborations.
Cost Allocation, Cooperative Game Theory, Collaborative Transportation, Row Generation, Vehicle Routing, Vehicle Routing with Profits, Column generation, Branch-price-and-cut, Algorithmic Capabilities