AI-Powered Remanufacturing

Sorting policies in remanufacturing systems

The quality of used products returned to recovery facilities is often highly uncertain. Quality grading and sorting policies are immediate solutions that are used in remanufacturing systems to handle this source of variability in incoming products. In this study, we offer a new sorting method based on both product’s internal factors such as future reusability of components, product identity data, and product health status as well as external factors such as market trends. The purpose of this study is to improve decision making in remanufacturing operations by integrating the product life cycle information, particularly product usage phase data, into determining both optimal sorting policies and End-of-Life/End-of-Use (EoL/EoU) decisions. 

Read more about this study here.

Multi-purpose disassembly sequence planning

Efficient disassembly operation is considered a promising approach toward waste reduction and End-of-Use (EOU) product recovery. However, many kinds of uncertainty exist during the product lifecycle which make disassembly decision a complicated process. The optimum disassembly sequence may vary at different milestones depending on the purpose of disassembly (repair, maintenance, reuse and recovery), product quality conditions and external factors such as consumer preference, and the market value of EOU components. In this study, we have applied a fuzzy method to quantify the probability that each feasible disassembly transition will be needed during the entire product lifecycle. 

Read more about this study here.

Uncertainty Management in Remanufacturing Systems

As market demand for remanufactured products increases and environmental legislation puts further enforcement on OEMs, remanufacturing is becoming an important business. However profitability of salvaging operations is still a challenge in remanufacturing industry. Several factors influence the cost effectiveness of remanufacturing operations, including uncertainties in the quantity of return flows and market demand as well as variability in the quality of received items. In this study, we have developed  a stochastic optimization model based on chance constrained programing to deal with these sources of uncertainties in take-back and inventory planning systems. The main purpose is to determine the best upgrade level for a received product with certain quality level with the aim of maximizing profit. 

Read more about this study here. 

Product Lifecycle Data towards Remanufacturing Decisions

The aim of this study is to incorporate the information collected from the product’s usage phase into making appropriate end-of-use and end-of-life (EOU/L) recovery decisions (e.g., reuse, remanufacturing, recycling, refurbishing, disposal) for consumer electronics. The analyses are focused on discovering a mutual link between the actual product usage behavior of consumers and the future reusability of devices. A data set of hard drive Self-Monitoring Analysis and Reporting Technology (S.M.A.R.T.) data has been analyzed and a reusability index for the components and the products has been presented. The results of the data analysis help identify different product clusters ready for remanufacturing and further develop decision models that match the best EOU recovery option with each product cluster with a heterogeneous quality level of incoming returns.