2022 Fiscal Year Research-status Report
Exact Optimization of Multiple Allocation Hub Location Routing Problem
| Project/Area Number |
20K04739
|
|---|
| Research Institution | Kyoto University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2020-04-01 – 2024-03-31
|
| Keywords | Multiple allocation / Hub location and Routing / Mathematical Modelling |
|---|
| Outline of Annual Research Achievements |
This research focuses on the multi-allocation hub location routing problem (MAHLRP), which arises in many-to-many logistics systems, such as express mail delivery systems. So far, the heuristics solution algorithm for the MAHLRP and an exact solution approach has been developed for a single allocation version of the problem, which will support the comparative analysis. The associated numerical analysis and the sensitivity analysis related with the exact solution algorithm of the single allocation problem was performed. The research article was submitted to a reputed international journal, and after a few rounds of reviews and revisions it was accepted and published.
An application of the MAHLRP in solving warehouse matching platform system (WMPS) has been developed. In the WMPS system provider connects businesses in need of warehouse space with companies that have excess warehouse capacity on its platform. The WSPS enables the usage of existing idle warehouses and avoid the inundation of factory-owned warehouses. As an MAHLRP application the WMPS can be setup where the factories and their final customers represent the non-hub nodes and the rental warehouses as hubs. It enables a dynamic supply network to respond to changing demand requirements both temporally and spatially.
The work on the exact solution of the MAHLRP has also progressed, where the algorithm has been coded and a basic numerical analysis is being carried out. Its performance and comparison with other algorithms and problems is still needed to show its advantages.
|
| Current Status of Research Progress |
Current Status of Research Progress
3: Progress in research has been slightly delayed.
Reason
Currently, the research on a new column generation based exact solution approach has been finalized. The MAHLRP is decomposed in the master problem and a subproblem. The subproblem is solved at each hub to generate promising allocations and routing. As non-hub nodes can be assigned to many hubs simultaneously, it effects the capacity utilization of the multiple chosen hubs. Therefore, a novel approach has been used in the master problem where the capacity constraints are kept in it, which are normally moved to the subproblem. In order to accelerate the process, subproblem is solved at the already open hubs in the previous iteration; once this process fails to deliver promising new columns, unopen hubs are tried. Its performance and comparison with other algorithms and problems is still in progress to show its advantages. For that additional numerical analysis would be required.
|
| Strategy for Future Research Activity |
Research papers describing the exact solution algorithm for the MAHLRP has been submitted to international conference along with a submission in an international journal. In order to show its advantage in addressing underlying logistics problem (such as the urban express mail delivery systems) comparison of the MAHLRP and its exact solution with other modeling techniques (such as multi-depot, simultaneous pickup and delivery problem) and their exact solution algorithm is required. In order to that, improvements at the level of the subproblem and at master problem of the newly developed exact solution algorithm will be incorporated based on the experts’ opinion and reviewers’ comments. It might also require more numerical analysis on benchmark instances and realistic case studies, which will be developed and solved accordingly. At the end of this academic year, it is expected that the research on the exact solution of MAHLRP will be published in some international journal.
|
| Causes of Carryover |
Due to COVID 19 pandemic, many international conferences were got cancelled and the chances of discussion on the proposed research were lost. In this academic year, the research outcomes will be shared with many researchers and scholars, as many academic conferences are returning to their normal arrangements, which would require travelling. These interaction will be key to confirm the research progress made so far and its publication and dissemination, also it will pave the way for future continuation of the research theme and future collaborations.
|
