A Simulation Study on Rider-order Matching in Meal Delivery Service

Abstract

Since the untact trend brought on by the COVID-19 pandemic, one of the fastest growing industries is the meal delivery industry. Unlike in the past, the meal delivery industry has been restructured to focus on ordering using delivery apps as the spread of mobile devices expands, and competition between the three major platform(Baemin, Yogiyo, Coupang Eats) is ongoing. A special turning point came in the competition between them in 2019, and that is “One order per trip” service introduced by Coupang Eats, a latecomer. This service was later introduced to other platforms due to high customer satisfaction, but it caused a problem in rider revenue under insufficient rider supply, causing a social issue. This study is based on the fact that this service fee does not take into account the number of deliveries, but only the simple delivery distance, and using simulation of rider-order matching, analyzes the correlation between the number of deliveries and the delivery distance and the imbalance in the delivery distance of riders. As a result of the simulation experiment conducted by fixing the number of riders and changing the number of orders, the ratio of the number of deliveries and the delivery distance is constant on average. And as the ratio of number of orders/number of riders increases, the standard deviation decreases. In addition, as the ratio of the number of orders to the number of passengers increases, the standard deviation of the travel distance decreases. This is also true when the locations of restaurants are concentrated, but in this case the absolute value of the standard deviation is smaller.