The paper co-authored with Maryam SteadieSeifi, Nico Dellaert and Wim Nuijten

A metaheuristic for the multimodal network flow problem with product quality preservation and empty repositioning

Just got published in the journal Transportation Research Part B: Methodological. The publisher was kind enough to give open access for 50 days. Click here to go directly to the paper.

Perishable product supply chains are international businesses and the horticultural industry of the Netherlands is an example. Everyday, a large number of cut flowers and plants, are transported around the world. These perishable products are fragile, have short shelf-lives, and travel long and in different climates. Temperature fluctuation and long handling time have a direct influence on their deterioration.

To ensure freshness of the products, they are traditionally transported by either air or road, and in temperature-controlled environment. However, their market is growing, and adding more air and road vehicles results in more expensive transportation, and causes various social and environmental issues such as congestion and pollution.

The challenge of finding the optimal transportation fleet plan is added to other operational issues such as resource management. An optimal transportation of perishable products needs synchronized flow with minimum waiting and handling. A cheap, diverse, flexible, and environmentally friendly transportation, ensuring freshness of the products while offering a competitive price, requires consolidation and switching from air and road to other modes of transport.

Finding the trade-off between minimizing operational costs and product quality preservation becomes an interesting research subject, which is the target of this paper.

In order to include product preservation requirements, we present a mode-space-time network where all types of multimodal operations such as holding, handling, transshipment, and transportation are included. Extra constraints include a product quality measure based on temperature and travel time, and enforce a maximum limit on the products after which the products are spoiled. Moreover, we integrate the forward flow of loaded RTIs with the backward flow of empty ones via a set of special constraints. We build upon the literature and propose an ALNS algorithm with new operators, improved scoring mechanism, and extra strategies, to solve this problem.