Under conditions of consumer panic buying, satisfying demand with the available products is a complex problem. In reality, most retailers accept alternative products during panic situations. This study considers the case of firm-driven substitution of products (differing in weight) based on retailer preferences over two periods. In the proposed model, panic behavior emerged in the first period and supply disruption occurred in the second period. Under this model, retail stores were segmented into high index (valuable) and low index (less valuable) customers. Before meeting the demand of low-index customers, wholesalers attempt to satiate high-index customer’s panic buying behavior. We determined the optimal number of units to be substituted, order quantities, and leftover units that generated maximum total profits for the wholesaler. The performance of the model was analyzed both with and without customer-segmented substitution. To gain managerial insights, we also examined the influence of both the degree of supply disruption and substitution costs on decisions and profits. The results can assist business managers to improve the decision-making process.
1 January 2022 in Anais da Academia Brasileira de Ciências
Managers are driven to accomplish significantly higher levels of operational performance due to the difficulty of today's dynamic production environment. Typically, the precision of production facilities and the efficiency of manufacturing systems are significant variables in productivity. Thus, predicting machine performance has become an inevitable challenge for production managers. However, the question of how managers can reliably assess the effectiveness of equipments for resource allocation remains unaddressed properly. This issue has received little attention in previous research, but it is important in today's manufacturing environment. This study introduces a hybrid moving average - adaptive neuro-fuzzy inference system (MA-ANFIS) to predict the possible effectiveness of equipment. Three real-world problems are considered when developing and evaluating three distinct equipment effectiveness prediction models. The evaluation confirms that the hybrid MA-ANFIS model based on Gaussian membership function outperforms other developed models. This comprehensive solution is packaged as a decision support system. This aids production managers in evaluating the equipment effectiveness, and effectively improving equipment's performance to reduce time and cost of bus body building.