Winter road maintenance operations involve a host of decision-making problems at the strategic, tactical, operational, and real-time levels. Those operations include spreading of chemicals and abrasives, snow plowing, loading snow into trucks, and hauling snow to disposal sites. We present a survey of optimization models and solution algorithms for the design of systems and the routing of vehicles for winter road maintenance. System design in winter road maintenance includes determining the level of service policy, partitioning the geographic region into sectors, locating the needed facilities (vehicle depots, materials storage facilities, and disposal sites), assigning the sectors obtained from the partitioning to various facilities, allocating contracts for various operations to private organizations, and replacing and sizing the vehicle fleet. We also present directions for future research.

We propose a hybrid method designed to solve a problem of dispatching and conflict-free routing of Automated Guided Vehicles (AGVs) in a Flexible Manufacturing System (FMS). This problem consists in the simultaneous assignment, scheduling and conflict-free routing of the vehicles. Our approach consists in a decomposition method where the master problem (scheduling) is modelled with Constraint Programming and the sub problem (conflict-free routing) with Mixed Integer Programming. Logic cuts are generated and used to prune optimal scheduling solutions whose routing plan exhibits conflicts. The hybrid method presented herein allowed to solve instances with up to six AGVs.

We describe a solution procedure for the Capacitated Arc Routing Problem with Refill Points (CARP-RP), problem inspired by the road network marking in Quebec. The CARP-RP deals with two different types of vehicles: the first type (called servicing vehicle - SV), which has a finite capacity, is used to service the arcs and the other (called refilling vehicle - RV) to replenish the first type of vehicle. The RV has a large capacity so it can meet the SV many times before returning to the depot. The problem consists of simultaneously determining the routes of both types of vehicles to minimize the total cost.

We develop a mathematical formulation and heuristics to solve a multi-period vehicle routing problem in a large scale production/distribution network. The routing must minimize inventory and travel costs while considering the latest delivery schedule (demands and due dates) of the retailers and the availability of the products (production schedule) and of the vehicles (heteregeneous fleet).