The Circular Economy and retail: using Deep Learning to predict business survival

Abstract

Abstract Background The Circular Economy system can improve the product cycle and changes the system and mentality, both for production and the consumer and has become a significant alternative to the classic economic model. The retail sector has also started to advance along these lines. Following an analysis of the state of the art of the Circular Economy and retailing, using bibliometric techniques, our research focuses on understanding if the relationship between circularity and retailing can help us determine a business’ survivability and resilience. To this end, data pertaining to 658 commercial premises from four cities were studied over a period of 11 years. A Deep Learning technique is applied using Long Short-Term Memory to determine if there is a relationship between the resistance of the selected commercial premises, their status in previous periods of time, the type of business activity, and their classification in the Circular Economy plane. Results The system predicts, on the set of tests, with a 93.17% accuracy, the survival of a commercial premises based on the activity, and circularity information before 2012. The results of the training also show very significant precision values of the order of 94.15% with data from the post-depression period. Conclusions The results show that businesses with activities related to the Circular Economy are more likely to survive over extended periods of time.