Multi-start Heuristics for Unit-Capacity Orienteering Problems

Abstract

We address a planning problem for the underground transit of goods with containers in mining companies. A problem instance contains tasks and capacities for their fulfillment. Tasks are transports of containers from their current locations to given destinations and deliveries and pickups of empty containers and materials. Besides, we address combinations of those tasks with precedence constraints. Storage locations provide a stock of empty containers and materials and are destinations for pickup containers and materials. Heterogeneous workers and a fleet of heterogeneous vehicles are available for performing the tasks. Each task has an assigned profit and can have a due date, which grants additional profit on observance. The objective is to maximize the benefit of fulfilled tasks and met due dates. We developed two heuristic solution approaches, creating tours either sequentially or in parallel. Furthermore, these constructive algorithms are extended to multi-start versions using randomization of vehicle and worker selections, container-type selection, and insertion criteria.