UNIST, Delivers 'Local Logistics' More Efficiently

Crowdsourcing Integrated Last Mile Delivery

A method to make logistics delivery more efficient has been developed.

The research team led by Professor Kwon Sang-jin from the Department of Industrial Engineering at UNIST (President Lee Yong-hoon) announced on the 11th that they have developed a logistics optimization methodology integrating crowdsourcing based on annual floating population data.

This study combines crowdsourcing with last-mile delivery, the final stage of delivering goods to customers, to determine the ideal terminal locations.

The newly developed methodology reduces total logistics costs by 3.09% and enhances the efficiency of logistics transportation through the coexistence of traditional logistics workers and crowd workers.

For last-mile delivery, the research team calculated employment and transportation costs that vary according to population density, rather than simply assigning crowd workers close to the final delivery destination. They comprehensively considered the transportation modes actually used by crowd workers and the carbon emissions associated with each mode.

The study integrated the emerging crowdsourcing-based logistics transportation into the existing logistics platform based on traditional logistics workers. This allows for more efficient delivery to the final customer while lowering overall logistics costs.

The research team applied the developed methodology to the Nam-gu area of Ulsan. They also analyzed the sensitivity of three cost parameters that could affect the efficiency of the last-mile delivery process (per-package fee, fuel price, and terminal setup cost). The results showed that the optimal location selection for hub terminals for logistics workers or crowd workers is not significantly affected by changes in per-package fees or fuel prices.

Furthermore, in areas with concentrated floating populations and sufficient supply of crowd workers, fluctuations in per-package prices did not affect the proportion of logistics assigned to crowd workers. This is because the logistics market responds relatively insensitively when logistics service supply is sufficient, even if various costs rise.

Conversely, in areas with low floating population density, price fluctuations such as per-package fees had a significant impact on the proportion of logistics assigned to crowd workers. When logistics service supply is insufficient, price increases such as wage hikes lead to a substantial rise in total logistics costs. The logistics market responds relatively sensitively in such cases.

In particular, the study revealed that when the cost required to install terminals increases beyond a certain level, it affects the optimal location selection of terminals regardless of floating population density. They discovered that when the installation cost ratio of ‘pickup stations’ for crowd workers exceeds 100% compared to the installation cost of ‘hub terminals’ for traditional logistics workers, the optimal number and locations of hub terminals and pickup stations change.

The research team also conducted additional studies on the strategic allocation of incentives to promote the introduction of electric vehicles in last-mile delivery, finding that this is more effective in reducing carbon emissions. By applying various traffic scenario simulations, they derived optimal logistics delivery solutions, laying the foundation for flexible responses in cases of traffic congestion.

Professor Sangjin Kwon (left) and first author Researcher Jaesung Kim

Jae-sung Kim, the first author and researcher, said, “This is a foundational study that serves as a stepping stone to drive innovative changes in last-mile delivery,” adding, “In particular, the new delivery options through crowdsourcing are expected to improve the efficiency of last-mile logistics and provide more eco-friendly logistics solutions.”

Professor Kwon Sang-jin stated, “This research comprehensively addresses the emerging issues of crowdsourcing, electric vehicles, and carbon neutrality in logistics,” and added, “We plan to conduct further studies integrating more diverse and complex future mobility technologies into logistics platforms.”

Supported by the Ministry of Science and ICT and the National Research Foundation of Korea, this paper was published online in January 2024 in Expert Systems with Applications and will appear in the August 2024 issue.

This content was produced with the assistance of AI translation services.