Idemitsu Kosan Co., Ltd (Idemitsu) and GRID Inc. (Grid) have announced that both companies are working on a PoC (Proof of Concept) project to automate and optimize ship routing using mathematical optimization and artificial intelligence methods such as machine learning and deep learning.
Although artificial intelligence technology has recently been applied to complicated tasks across many sectors, the supply chain and logistics sectors still rely heavily on the knowledge of experienced workers.
This is the same in the refinery industry. The strong dependency of ship routing on the knowledge of experienced operators makes automation a difficult task, due to complex restrictions on the combinations of vessels and cargo ships in ports and to dynamic supply, demand and weather conditions.
As a result, previous efforts to apply mathematical optimization methods to automation achieved limited success.
As an unprecedented effort in this industry, Grid, an AI tech start-up based in Japan focusing on the industrial sector, and Idemitsu, one of the leading refinery companies in Japan with significant expertise in ship operations, jointly aim for autonomous optimization of ship routing by combining their expertise, along with support from Mitsui & Co ., Ltd, which has deep knowledge in both artificial intelligence and supply chain management.
Idemitsu and Grid began the PoC project in June 2019 and have confirmed that AI-based ship routing already exceeds past human-based results in various KPIs, such as efficiency and inventory rate.
Idemitsu and Grid will continue the project until May 2020, with further analysis including the application of deep reinforcement learning methods, a ground-breaking effort in the supply chain sector.
Ultimately Idemitsu and GRID aim to apply AI to ship routing to optimize the entire supply chain.