PERIODIC ROUTING OF SERVICE VESSELS WITH TIME WINDOWS IN OFFSHORE OIL INDUSTRY

Service vessels in the offshore oil industry have a noticeable logistics role in protection and continuity of quality and quantity of oil production. Offshore oil production takes place on oil wellhead platforms. In order to maintain oil production in offshore areas, using supportive vessels and mobile wellhead vessels, the National Iranian Oil Company periodically provides public services and well services for equipment located on oil wellhead platforms. However, in the meantime, mobile wellhead servants themselves need to receive service from supportive vessels in specific periods and, also, specific time windows. In fact, mobile wellhead servants have a dual role. On the one hand, they have a role to play in providing well services for oil wellhead platforms and, on the other hand, they have the role of receiving public services from supportive vessels. Service vessels are few in number compared to service requirements; however, they have a high shipping cost instead. Therefore, on the one side, using them without an appropriate plan leads to significant lost opportunity cost of oil production due to service shortages and, on the other hand, imposes high shipping costs. In order to determine the periodic routing of service vessels, a mixed-integer programming model is proposed with the aim of minimizing the cost of shipping and the lack of services. Given that the mathematical model is Np-hard, a heuristic algorithm based on practical conditions of the problem is introduced to solve the problem, and its performance is evaluated using real-life instances related to different oilfields located on the Iranian side in the Persian Gulf. Numerical results indicate the algorithm's speed in achieving appropriate solutions in a very short amount of time compared to existing algorithms. In fact, implementing the heuristic algorithm leads us to creating very good solutions in the primary iterations of running the heuristic algorithm and improving them in the next iterations.