Islam Md Tariqul, Azeem Abdullahil, Jabir Masum, Paul Ananna, Paul Sanjoy Kumar
Department of Industrial and Production Engineering, Bangladesh University of Engineering and Technology, Dhaka, Bangladesh.
Faculty of Engineering and Information Technology, University of Technology Sydney, Sydney, Australia.
Ann Oper Res. 2022;315(2):1703-1728. doi: 10.1007/s10479-020-03639-z. Epub 2020 May 15.
This study develops an inventory model to solve the problems of supply uncertainty in response to demand which follows a Poisson distribution. A positive aspect of this model is the consideration of random inventory, delivery capacities and supplier's reliability. Additionally, we assume supplier capacity follows an exponential distribution. This inventory model addresses the problem of a manufacturer having an imperfect production system with single supplier and single retailer and considers the quantity of product (Q), reorder points (r) and reliability factors (n) as the decision variables. The main contribution of our study is that we consider supplier may not be able to deliver the exact amount all the time a manufacturer needed. We also consider that the demand and the time interval between successive availability and unavailability of supplier and retailer follows a probability distribution. We use a genetic algorithm to find the optimal solution and compare the results with those obtained from simulated annealing algorithm. Findings reveal the optimal value of the decision variables to maximize the average profit in each cycle. Moreover, a sensitivity analysis was carried out to increase the understanding of the developed model. The methodology used in this study will help manufacturers to have a better understanding of the situation through the joint consideration of disruption of both the supplier and retailer integrated with random capacity and reliability.
本研究开发了一种库存模型，以解决响应遵循泊松分布的需求时的供应不确定性问题。该模型的一个积极方面是考虑了随机库存、交付能力和供应商的可靠性。此外，我们假设供应商能力遵循指数分布。此库存模型解决了制造商在生产系统不完善且只有单一供应商和单一零售商的情况下的问题，并将产品数量（Q）、再订购点（r）和可靠性因素（n）作为决策变量。我们研究的主要贡献在于，我们考虑到供应商可能无法始终交付制造商所需的精确数量。我们还考虑到需求以及供应商和零售商连续供应和缺货之间的时间间隔遵循概率分布。我们使用遗传算法来寻找最优解，并将结果与通过模拟退火算法获得的结果进行比较。研究结果揭示了决策变量的最优值，以在每个周期内最大化平均利润。此外，还进行了敏感性分析，以加深对所开发模型的理解。本研究中使用的方法将帮助制造商通过综合考虑供应商和零售商的中断情况以及随机能力和可靠性，更好地了解实际情况。
