نوع مقاله : مقاله پژوهشی
نویسندگان
1 استادیار، گروه مهندسی صنایع، دانشکده و پژوهشکده فنی و مهندسی، دانشگاه جامع امام حسین (ع)، تهران، ایران.
2 دانشجوی دکتری، گروه مهندسی صنایع، دانشکده و پژوهشکده فنی و مهندسی، دانشگاه جامع امام حسین (ع)، تهران، ایران.
3 کارشناس ارشد، گروه مهندسی صنایع، دانشکده و پژوهشکده فنی و مهندسی، دانشگاه جامع امام حسین (ع)، تهران، ایران.
چکیده
کلیدواژه‌ها
موضوعات
عنوان مقاله [English]
نویسندگان [English]
Introduction and objectives: Refurbishing an oil rig is one of the best alternative options for renting this capital equipment. Because the cost of renting it is tens of thousands of dollars per day. This research deals with the scheduling of activities and inventory planning of the supply chain of the Sina 1 drilling rig reconstruction project. Since there is no record of doing such work in the country, project scheduling is done in the conditions of uncertainty of activities. Also, the uncertainty of the implementation time of the activities itself will increase the understanding of the project owners about the scheduling of the activities and will open a wider view of the project and the upcoming activities. In this model, the costs of the contractor and suppliers are seen at the same time, and the non-deterministic scheduling of activities and planning of orders is done in such a way that the total costs of the entire chain are minimized.
Method: According to the number of variables and limitations of the mathematical model for scheduling the activities of the supply chain of the drilling rig reconstruction project, the investigated problem is a type of complex NP-hard problem. Therefore, meta-heuristic methods are used to solve such problems, which produce near-optimal solutions in a reasonable time compared to exact methods. In this research, electromagnetic algorithm is used to solve this problem. The mentioned algorithm has been used in a real project (Leg section of Sina 1 drilling rig).
Findings: The mathematical model presented in this research has been coded using the electromagnetic algorithm in the MATLAB software environment, and the input parameters of the problem include general parameters and control parameters of the electromagnetic meta-heuristic algorithm. The general parameters are related to the specifications of the contractor, supplier, activities, resources and consumables. In order to prove the efficiency and effectiveness of the designed electromagnetic algorithm, three problems were selected. The first problem has five activities, the second problem is actually the same five activities as the first problem but with two suppliers. The third problem was selected with large dimensions. First, the exact solution for the problem was obtained using AIMMS software, then the solutions and the solution time obtained from the electromagnetic algorithm were compared with it. It should be noted that as the dimensions of the problem increase, the time to reach the solution in AIMMS software increases significantly (exponentially). The proposed electromagnetic algorithm shows an acceptable performance according to the solution time. The answer time of the electromagnetism meta-heuristic algorithm to solve the sample problem is 115 seconds, but the time to reach the answer by the exact AIMMS software is about 747 seconds. Also, the relative difference of the meta-heuristic algorithm of electromagnetics compared to the exact software of AIMMS is almost one percent. Finally, a sensitivity analysis was used to analyze the impact of different payment methods on the project's supply chain cost.
Conclusion: In this research, the modeling and solution of the two-level project supply chain including the contractor and suppliers are presented, and the costs related to the project and inventory for both members of the chain are minimized. By comparing the solution results of the exact method and the electromagnetic algorithm, it was found that the time of solving with the electromagnetic algorithm is much less than the time of solving the exact method. Meanwhile, the relative difference in the quality of the answer is almost one percent. These cases show the effectiveness and efficiency of the proposed algorithm. The obtained results show that the proposed electromagnetic algorithm for this problem is an effective algorithm and converges to the optimal solution.
کلیدواژه‌ها [English]
)