Baemin’s ‘AI Recommended Dispatch’ Reduces Rider Accident Probability by 27.8%

National University of Singapore Research Team “AI Recommended Dispatch Proves Safety and Income Effects”

A study found that Baedal Minjok (Baemin)’s ‘AI recommended dispatch’ reduced the accident rate of delivery riders by 27.8%. The analysis is that AI recommended dispatch needs to be spread throughout the industry instead of competitive dispatch, which is called a ‘battle call’, as it enhances rider safety and increases income opportunities.

According to the elegant young people in charge of Baemin’s logistics service on the 16th, At the summer workshop hosted by the Korea Chapter of the Association for Information Systems (KrAIS) at Seoul National University, Professor Kyung Na-kyung of the Department of Computer Science at the National University of Singapore revealed the safety-enhancing effects of the AI ​​recommendation dispatch system through research titled, “Is the algorithm a guardian?: The impact of the AI ​​recommendation dispatch system on the safety of delivery riders.”

Baemin Connect Rider

View Original Icon

The research team analyzed the operation and accident data of riders who deliver through the Baemin Connect app through Elegant Youth. They compared and analyzed the accident patterns of riders who used AI-recommended dispatching and riders who did not use it during the same period. As a result, the accident probability of the rider group using AI-recommended dispatch was 27.8% lower than that of the rider group not using it. If the number of monthly accidents among 100 riders using the competitive dispatch method is 10, this number decreases to 7.2 for the rider group using AI-recommended dispatch.

The reason why AI recommended dispatch can greatly reduce the probability of accidents is because it creates an environment where riders can focus on driving by greatly reducing factors that can distract them compared to the competitive dispatch method commonly used in the delivery industry. The competitive dispatch method exposes orders to multiple riders at the same time, and the rider who accepts first wins the dispatch. In order to obtain dispatches, riders have no choice but to frequently check their smartphones while driving, and as they cannot focus on driving itself, the risk of accidents naturally increases.

The AI ​​recommendation dispatch system introduced by Baemin in 2020 suggests optimal dispatches to each rider based on various information such as the rider’s location and delivery status. In particular, by only exposing dispatches to the rider in question while deciding whether to accept the dispatch, it prevents situations where riders compete for the same dispatch with other riders, reducing the need for riders to check their smartphones while driving.

The research team explained, “AI recommended dispatching has been proven to be effective in improving riders’ income in previous studies, and this study revealed that in addition to income, it creates an environment where riders can focus on driving, which significantly reduces the probability of accidents.” Professor Kyung Na-kyung emphasized, “Although many delivery companies are still using the combat call method, the introduction of AI recommended dispatching should be expanded throughout the industry for various aspects such as rider safety and income.”


Reporter Kim Chul-hyun kch@asiae.co.kr

<ⓒ투자가를 위한 경제콘텐츠 플랫폼, 아시아경제(www.asiae.co.kr) 무단전재 배포금지>

Source: www.asiae.co.kr