A trilevel r-interdiction selective multi-depot vehicle routing problem with depot protection

Area of Science:

• Operations Research
• Supply Chain Management
• Optimization Theory

Background:

• Critical facilities are essential components of supply chain networks, necessitating protection strategies.
• Existing research lacks comprehensive models for safeguarding vital depots against intelligent adversaries in routing networks.

Purpose of the Study:

• To develop a trilevel optimization model for the protection of critical depots in a routing network against an intelligent adversary.
• To formulate the problem as a defender-attacker-defender game, termed the trilevel r-interdiction selective multi-depot vehicle routing problem (3LRI-SMDVRP).

Main Methods:

• A smart exhaustive enumeration approach is employed for the upper and middle level problems (ULP and MLP).
• A metaheuristic algorithm, hybridizing Variable Neighborhood Descent and Tabu Search, is designed for the lower level problem (LLP).
• The model is tested on existing multi-depot vehicle routing problem (MDVRP) and selective MDVRP (SMDVRP) benchmark instances.

Main Results:

• The proposed comprehensive method demonstrates effectiveness in addressing the defender-attacker-defender game.
• Numerical experiments on a large number of 3LRI-SMDVRP instances validate the algorithm's performance.
• The study provides a robust solution for optimizing vehicle routes and customer service post-attack.

Conclusions:

• The trilevel optimization model and associated algorithms offer a significant advancement in securing critical supply chain facilities.
• The research provides valuable insights for decision-makers in managing supply chain disruptions and optimizing routing strategies under adversarial conditions.