Combined Monte Carlo simulation and memetic algorithm for a stochastic multi-trip inventory routing problem

The present paper introduced, modelled, and solved a real-life logistics problem arising in the hospital complex of Rabat (Morocco). To manage efficiently the distribution process, hospital managers propose to adopt a vendor managed inventory system that involves the establishment of a central pharmacy, which will control the inventories of hospitals. The central pharmacy will also be responsible for delivering full roll containers of medicines to hospitals and the collection of empty ones for reuse in the next period. All deliveries and returns are performed simultaneously by a homogeneous fleet of vehicles that may perform several trips per period. Each hospital defines a time window wherein the service can begin. This problem is known in the literature as the multi-trip inventory routing problem with time windows and simultaneous pickup and delivery. Given the complexity of the problem, we propose a combined simulation-heuristic problem, which combines Monte Carlo simulation and memetic algorithm. The performance and effectiveness of the proposed approach are clarified by computational experiments.