AVEVA, a global leader in industrial software, driving digital transformation and sustainability, announced the successful deployment of AVEVA Schedule AI Assistant at Hokkaido Refinery, operated by Japanese energy company Idemitsu Kosan Co., Ltd.
AVEVA Schedule AI Assistant, part of AVEVA Unified Supply Chain, is an optimization and artificial intelligence (AI)-infused cloud-based solution. Designed to empower crude oil operational schedulers at refineries to explore and rank possible scenarios for efficiency, profitability and emissions, the solution helps to accomplish days’ worth of work in seconds, enhancing business agility and optimizing decision-making. Analytics automatically generate and evaluate multiple schedules and anticipated events within the complete supply chain. The AI-powered solution then recommends the crude scheduling strategy that best fits the organization’s safety, sustainability and economic value chain objectives.
Petrochemical refineries aim to optimize their planning and scheduling processes to achieve greater productivity and stabilize production flows. Through schedule optimization, management teams can take charge and prepare for changing environments: A production scheduler plans the flow of crude oil throughout a refinery – unloading it from vessels, transferring to storage tanks, preparing the charging schedule for different crude oil distillation units, considering restrictions on units’ capacity, flow and compositions, carrying out tank maintenances, and managing the production of end-product to maximize profitability on the spot market. As such, efficiency is limited by human fallibility and time constraints. The multi-step process and numerous variables can lead to errors, making it challenging to overall optimization.
By leveraging the predictive and prescriptive artificial intelligence in AVEVA Unified Supply Chain, customers can seamlessly drive scheduling decisions, capitalize on economic opportunities and unlock more value with AVEVA Schedule AI Assistant.
Idemitsu Kosan had already transformed its refinery operations by implementing AVEVA Unified Supply Chain to optimize its enterprise value chain and standardize its combined downstream operations. The company had introduced AVEVA Unified Supply Chain to develop planning and scheduling models for enhancing decision-making process for higher refinery margins, optimizing refinery operations and minimizing cost of crude oil and transportation. Having realized significant benefits from their existing AVEVA solution, deploying AVEVA Schedule AI Assistant at their Hokkaido refinery was the natural next step for Idemitsu Kosan in their digitalization program using AVEVA Value Chain Optimization .
“With the Schedule AI Assistant, AVEVA Unified Supply Chain solution is transforming operational schedulers into value chain strategists by presenting many possible solutions and placing them at the center of a cloud and AI-assisted supply chain decision support system that recommends the most efficient course of action, saving teams time and enabling them to exceed their objectives. AVEVA’s comprehensive software portfolio gives our clients the highest levels of efficiency and enables true value chain optimization,” said
Harpreet Gulati, Senior Vice President, Planning, Simulation and Optimization Business.
The Schedule AI Assistant elevates AVEVA Unified Supply Chain solution to the next level to further extend its capability of addressing each component of the value chain, right from feedstock data management, trading, production planning and network optimization, to scheduling and performance monitoring.
Seiji Yoshii, Manager, System Development＆ Maintenance Section No2 Information Systems Department, Idemitsu Kosan said,“Initially, we felt very difficult to execute this project technically. However, we developed the software for practical use and put it into operation for actual inspection in a year. We are confident that this achievement will result in a new generation of schedulers that will significantly change the traditional concept of try-and-error simulation-based scheduling. we look forward to initiating further developments and lead to great results.”
