ارائه ي يك مدل موجودي براي خرابي غير آني كالا در يك زنجيره تامين دوسطحي

An inventory model for noninstantaneous deterioration items in a two‌echelon Supply chain

در اين تحقيق يك مدل رياضي براي زنجيره‌تامين دوسطحي متشكل از يك خريدار و يك توليدكننده براي يك سيستم يكپارچه موجودي با خرابي غير آني اقلام ارائه مي شود كه تقاضا احتمالي است و از توزيع نرمال پيروي مي كند. فرض مي شود خرابي از توزيع وايبول سه پارامتر پيروي مي كند. در شرايط واقعي در نظر گرفتن كمبود هم به‌صورت پس افت و هم به‌صورت فروش از دست رفته الزامي است، بنابراين هر دو نوع كمبود نيز در مدل بكار گرفته مي شود. هدف از اين مدل تعيين سياست بهينه سفارش دهي است بطوري كه مجموع هزينه هاي زنجيره تامين كمينه گردد. براي حل مدل از الگوريتم ژنتيك بهره برده شده است. همچنين جهت تحليل مدل و بررسي تأثير برخي از پارامترهاي مهم و تأثيرگذار بر جواب بهينه مدل موجودي از تحليل حساسيت نرخ خرابي و سطح اطمينان استفاده مي‌نماييم. در نهايت مقدار بهينه هزينه مورد انتظار زنجيره تامين تحت تصميم گيري يكپارچه و غير يكپارچه تعيين و مقايسه مي گردد.

Most of the inventory control models assume that items can be stored indefinitely to meet the future demands. However, certain types of commodities either deteriorate or become obsolete in the course of time and hence are unstable. In this study, a mathematical model is presented for a twoechelon supply chain including a buyer and a producer for an inventory integrated system with noninstantaneous of items that demand is probable and follows a normal distribution. Since, the rate of deterioration describes the condition deterioration the goods and regarding the relation between time and deterioration rate is probable rather than the fixed rate of deterioration. In reality, considering the shortages is necessary in both forms of backlogging and lost sales. Therefore, both kinds of shortages are used in the model.    The main goal of this model is determining the optimal ordering policy so that the total cost of supply chain is minimized. The proposed model is solved for some problems by Lingo software. The validity of model is determined by sensitive analysis and the problem is known a NPhard one, hence a genetic algorithm has been used in order to solve the model problem. The rates of deterioration and confidence level sensitivity analysis have been applied to analyze effect of some important parameters affecting on optimal solution of the inventory model.    Finally, the optimal value of the expected cost of supply chain under integrated and nonintegrated decisionmaking has been determined and compared. The results show the efficiency of algorithm