Using AI in Manufacturing to Eliminate Late Production Orders – Case study of OTD Improvement in Hi-Mix/Lo-Volume Production Environment

Attend this session to learn how a complex manufacturer significantly reduced lateness in production by implementing real-time production priorities and predictive analytics to give production the information they need to make the right decisions that effectively:

• reduce WIP (visibility of the right time to start workorders),

• eliminated material chasing, (do we have the material to get started?), and

• ensure the right job is being worked on at every workorder to maximize flow and ensure on-time delivery.

Further, after ensuring production was effectively execution priorities, the company employs predictive analytics by creating a digital twin of its future production environment.  This powerful simulator shows future Load v Capacity, identifies production bottlenecks (people, machines, tooling, materials), and accurately predicts the expected finish date of every order. 

The digital twin then serves as a foundation to perform powerful “What-if” analysis by inputting proposed changes in capacity, resources, load (new orders or quotes), and see the impact on specific or overall OnTime Delivery (OTD) and Revenue/Contribution Margin.

Finally, as the software being used is AI-compliant/accessible/enabled, we will show LLM capabilities to access key scheduling information such as, “We have a new prospect who would like 100 of the same part we did for the ACME A8000 order.  If they place the order by Monday, when can we promise delivery given our current backlog?