User Experience Study on Self-Checkout System of Hypermarkets in Taiwan

In recent years, many hypermarkets have introduced self-checkout technology. The introduction of new technology is meant to provide a better experience for consumers and reduce the workload of store staff. However, the fact that a few extra staff members are still required to assist consumers in the self-checkout process reflects that consumers still have doubts about the service quality.In this study, 30 subjects who had never used self-checkout were invited to conduct task experiments on the machines at the two large hypermarkets in Taiwan. The assessments included (1) Functional Sequence: to understand the difference between the expected operation process and the actual one, (2) Task Operations: including membership login, scanning of different packaged products, electronic receipt registration, and payment checkout, with the operating time recorded throughout the experiment, (3) to understand the difference between the two self-checkout Systems in terms of interface functions and evaluation by using System Usability Scale (SUS), NASA-TLX and semi-structured interview.It was found that incorrect operation of the payment devices occurred in both machines of the two hypermarkets. The prompts from the system interface were not effective in helping the subjects to operate the machines smoothly. We found that (1) The interface of the hypermarket A was laid out with both illustrations and function buttons, resulting in 73% of the subjects failing to distinguish the difference between the two types of information and causing incorrect operation and failure to find the device; (2) The interface of the B market uses the photo prompt information; 60% of the test subjects ignore the photo, and the interface information cannot effectively assist the test subjects to operate; (3) The design of the machine uses a vertical screen (24" for both machines) and a full-page layout interface, which makes it easy for the test subjects to ignore the buttons, prompts, and hardware devices located in the corners of the interface when operating at close range.Moreover, the SUS of the machines in the two hypermarkets were 74.66 (C level) and 51.83 (F level) respectively, and neither of them could meet the operational requirements. This study found that when the test subjects operated the machines for the first time and faced with the queue of people, their mental stress level and error rate increased. The future interface design of the system should strive for precise operation and information understanding within a short period of time.