Document Type



This study attempts to determine the optimal goods ordering periods for internet stores by considering time-dependent consumer demands and close demand-supply interactions. In order to capture dynamic and time-sensitive consumers, the entire study period is divided into a number of ordering periods with various duration. In the demand side, the study formulates a consumer utility function to construct a binary logit model, which determines consumers’ choice probabilities between internet shopping and conventional in-store shopping. The expected choice probability of choosing internet shopping is aggregated by a transformation probability density function of individual income based on the logit model. Then, the study further aggregates individual consumer choice probability to estimate the total demand for internet stores by considering variations in access time to retail stores, and delay time of receiving ordered goods. In the supply side, the study formulates transportation costs considering extra labor cost due to 24 hours business hours of internet stores, and constructs inventory costs reflecting the relationship between the batch ordering of goods, which are made by internet store operators to their suppliers and continuous ordering of goods, which are made by their consumers. Finally, a case study and sensitivity analysis are provided by R-company in Taiwan to illustrate the application of the models. The results show how the operators of internet stores should determine the number and duration of ordering periods in response to time-dependent consumer demand, thereby maximizing their profits.