Narrow Results

The daily responsibilities of e-commerce enterprises, such as operational preparation and market assessment, are greatly impacted by sales forecasting. Since the market is constantly shifting, e-commerce enterprises must immediately solve the important difficulty of effectively forecasting sales. Traditional forecasting techniques also have limitations. This study offers a novel fuzzy logic-based method for improving sales forecasting’s handling of uncertainty. This study gathers a wide range of data from multiple sources, such as transaction logs, analytics of consumer behavior, and market trends. To guarantee its quality and dependability, the gathered data are pre-processed by using min–max normalization and glowworm swarm optimization (GSO) was employed for feature selection. This research develops a fuzzy logic multi-criteria (FLMC) algorithm that takes into account a number of influencing elements, such as customer behavior, customer demographics, market trends, website traffic and engagement, product availability, and seasonal fluctuations, by leveraging fuzzy logic’s ability to simulate imprecision and ambiguity. The suggested algorithm uses a multi-criteria decision-making process to assess and evaluate these variables, producing useful information that helps firms maximize their marketing, pricing, and inventory control initiatives. The efficacy of the FLMC algorithm in enhancing forecasting accuracy is demonstrated empirically. The findings demonstrate how the algorithm could be used to mitigate the unpredictability found in e-commerce settings, promoting better decision-making and increasing operational effectiveness.