Determining the quality index of online shopping websites by using fuzzy logic

Abstract

Online shopping is a platform of e-commerce allowing consumers to purchase goods or services directly from a seller over the Internet using a web browser or a mobile application. Due to Covid-19 pandemic, almost every person has started to spend time at home and preferred e-shopping instead of physical shopping. A practical, applicable, easily updated and effective measurement of the quality of e-shopping sites can make a positive contribution both for customers to make the right choice and for companies to evaluate their positions. In this study, we develop an inference system in fuzzy environment to quantitatively describe the service quality of online shopping websites. The proposed system consists of four sub-fuzzy and one main Mamdani-type inference systems. The inputs for the main fuzzy inference system are considered as cargo and delivery performance, customer support and campaign performance, ease of payment and return, interface quality, and privacy-security. On the other hand, the first sub-fuzzy system, whose inputs are Quality of Shipping the Products, Quality of Order Tracking System, and Delivery Speed of the Products, provides inferences for cargo and delivery performance. The second sub-system is formed for customer support and campaign performance where there are two inputs. The third sub-system gives similarly an output about the ease of payment and return with two inputs. In the fourth system, evaluation of the interface quality of e-shopping sites is carried out, which includes four inputs. Finally, a quality index for online shopping website is determined by entering the values of sub-fuzzy systems and privacy-security to the main fuzzy inference system. In addition, a survey is applied for verifying and supporting the study. In this survey, information about preferred online shopping sites is requested in order to obtain the input values in fuzzy systems.