When the chemical division of SCG needed to solve a complex manufacturing problem, its leaders made a bold choice: collaborate with Microsoft and use artificial intelligence to supplement its own innovation.
Why choose a technology company to help with a chemistry problem?
“Our problem was relatively complex, due to the nature of chemical processes,” said Surachate Chalothorn, CTO for Olefins and Operations of Chemicals Business, SCG. “To achieve our goal for operational excellence, having a good simulation model was not enough. We believed that chemical process simulation needed to integrate with AI technology to solve our complex problems.”
The solution the company chose was Project Bonsai, the first service component of Microsoft’s autonomous systems vision. The Microsoft investment in autonomous systems technologies is designed to empower traditional engineers, like the chemical engineers who work at SCG, to use AI to find new solutions to complex problems.
“We worked together to tackle this challenge,” said Yanon Lorpatarapong, AI Lead Engineer for Chemicals Business, SCG. “It wasn’t just buying a solution. Our team actually did the coding and contributed a lot during the project.”
Project Bonsai, now in public preview , as announced at Microsoft Build , is a machine teaching service to create and optimize intelligence for industrial control systems. Through machine teaching , subject matter experts without an AI background can break down their expertise into steps and tasks that are imparted to AI agents. The experts can specify desired outcomes and criteria, then supervise the AI agents as they work to solve problems in simulated virtual environments. The experts provide feedback and guidance that trains the AI agent to dynamically adapt within the simulation. Once they are sufficiently trained in the simulation, those AI agents can use their knowledge to power autonomous systems in real-world applications.
Traditional engineers and software developers alike can use the Project Bonsai platform to create custom solutions to their specific problems.
“This approach bridges AI science and software to the traditional engineering world, enabling fields such as chemical and mechanical engineering to build smarter, more capable and more efficient systems by augmenting their own expertise with AI capabilities,” said Mark Hammond, Microsoft general manager for autonomous systems.
Since the initial announcement of Project Bonsai, Microsoft partnered with customers on a broad range of projects within manufacturing, process control, robotics, energy, smart buildings, consumer durable goods and other sectors. Those projects informed documentation, templates and training videos that people can now access in a self-serve fashion.
“We are opening it publicly so that anyone who wants to come in and give it a spin can do so,” Hammond said. “You can come in and take advantage of the demonstration hardware, simulations and partnerships we’ve put together. You’ll be able to understand not just the theory behind how a system can help you, but actually try it yourself and find out whether it can help solve your problems directly.”