*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
The focus of this dissertation are problems of fleet management in freight transportation systems. Effective management requires both effective planning and control decisions. Plans are often generated using estimates of how the system will evolve in the future; during execution, control decisions need to be made to account for differences between actual realizations and estimates. The benefits of plans designed to satisfy customer demands at minimum cost can be negated by performing costly adjustments during the operational phase. Plans should prepare the system appropriately for the uncertain conditions to be faced during future operation. A planning approach that permits effective control during execution is proposed in this dissertation. This approach is inspired by recent work in robust optimization.
We develop concepts and methodologies that are applied in two fundamental problem areas: (i) dynamic asset management and (ii) vehicle routing.
In practice, the fleet management planning and control problem is usually decomposed in two parts; the problem of repositioning empty units to account for flow imbalances, and the problem of allocating units to customer demands and handling the corresponding loaded moves. An alternative integrated dynamic model for asset management problems is formulated as large-scale multicomodity flow problem in a time-discretized network. A computational study provides evidence that operating costs and fleet sizes may be significantly reduced with the integrated approach. However, results also illustrate how a model that does not consider the inherent demand uncertainty generates fragile plans in which serving realized customer demand can be either impossible or very costly. We then propose a planning approach for the empty repositioning problem that incorporates demand and supply uncertainty at every period of the planning horizon using interval around nominal forecasted parameters. The intervals define the uncertainty space for which buffers need to be built into the plan in order to make it a robust plan. A plan is said to be robust with respect to a deviation from forecasted nominal values if there exists a set of feasible control decisions drawn from a restricted control space. Computational evidence suggests that this approach is tractable.
The traditional approach to address the Vehicle Routing Problem with Stochastic Demands (VRPSD) is through expectation minimization of the cost function. Although this approach is useful for building fixed routes with low expected cost, it does not directly consider the maximum potential cost that a vehicle might incur when traversing the tour. Our approach aims at minimizing the maximum cost. Computational experiments show that our robust optimization approach generates solutions with expected costs that compare favorably to those obtained with the traditional approach, but also that perform better in worst-case scenarios. We also show how the techniques developed for this problem can be used to address the VRPSD with duration constraints.
