Designing a New Model for Producing LED Lamps Considering the Quality Effects of Raw Materials on the Rate of Returned Items

Supply chain management has always been of interest to researchers because of generating competitive advantage, such as reducing costs and increasing customer service levels. However, the mathematical models that can take into account the specific conditions of today’s supply chain are negligible. The impact of factors such as the level of production quality on the returned items rate is one of the subjects that have received less attention. In this research, the mathematical model of the closed-loop supply chain developed concerning supplier selection and the effects of the quality level of provided raw material into the returned items rate

The investigated supply chain is a two-level closed-loop supply chain modelled with a multi-period time horizon. If there are defective products, customers can deliver the products to the collection centres and receive new products. In each period, the returned items rate depends on the level of production quality, and the level of production quality depends on the quality of raw materials provided by suppliers; thus, the returned items rate is indirectly affected by the quality of raw materials. Such is a subject that was not studied earlier.

Case study represents a real example of production and recycling of LED lamps. By comparing the financial data announced by the company with the results obtained from the model, the model showed a 3.37% improvement in the objective function. In the current situation, all materials purchased from only one supplier at quality level B (out of the three quality levels A, B and C); in the optimal case, however, most of the purchased materials are at quality level A.

 The relationship between quality level and the return rate is recognized by considering nonlinear functions, resulting from the fitting of these functions with real data. Also, the assumptions of multi-products in the supply chain and lateral transshipments between distributors/suppliers suggested for further study. In the case of considering multi-products, different selling prices are reasonable for the products. Also, greater profitability is achievable by considering coordination among the members of the supply chain. 
 Practical implications: As practical implications, manufacturers can use this optimization model to i) determine the amount and quality level of raw materials provided by each supplier at each period; ii) determine the amount and quality of the final product produced at each period; iii) specify the number of distributors' orders at each period as well as the number of backorders; iv) determine the amount of returned items at each period. Thus, solving the model provides a comprehensive plan to the manufacturer in various dimensions of production.

 This theoretical-practical research performed to investigate one of the main problems of production systems. Unfortunately, in recent years, critical economic conditions, sanctions, and currency fluctuations have caused suppliers to provide raw materials with lower qualities instead of materials with high quality. These low-level materials also have an impact on the final products during production and reduce their quality level. As a result, the return rate of defective items increases and consequently, the loss of the company increases. Therefore, like what happens in the real world, the manufacturing company has to choose a combination of the materials with different qualities from the suppliers to maximize the company's profit. Mathematical modelling is needed to solve such a problem, and current research is an attempt to reach this mathematical model by considering all aspects of the production process.

The main novelty of this research is considering the quality levels for raw materials and its indirect effect on the return rate of defective items; so that the quality level of raw materials determines the quality level of final products, and the quality level of final products affects the returns rate. For this purpose, an integer programming method developed to formulate and solve the model.