Optimizing cobots inspection routing in chemical plants

Maintaining core infrastructure, such as high-pressure piping and storage tanks, is essential for ensuring the safe operation of chemical plants. Although recent technological advancements have broadened the scope of robot application in logistics and infrastructure surveillance, robots cannot yet fully replace human inspectors in chemical plants due to limitations imposed by complex operational environments and cognitively demanding tasks. In the context of human-robot collaboration (cobots), this paper presents a location-routing problem with cobots, in which robots are responsible for high-risk, repetitive tasks, while humans focus on tasks requiring complex physical access, cognitive interpretation, and hands-on intervention. The objective is to minimize the total cost, which comprises the depot opening cost, the fixed utilization cost of the employed cobots, and the total travel cost, by simultaneously determining which depots to open and how to route the cobots. The problem is formulated as a mixed-integer linear programming model, and a hybrid evolutionary algorithm featuring a backbone-based crossover operator is proposed to solve large-sized instances. Extensive experimental results on benchmark instances show that our algorithm consistently finds solutions comparable to or better than those obtained using Cplex. Finally, a case study confirms the efficiency of the algorithm and highlights the cost-effectiveness of cobots in reducing operational costs while maintaining safety in chemical plants.