Skip to content

Development of On-Time.ai software

The need for On-Time.ai software became apparent from the George Group’s 20 years of experience with Lean Six Sigma implementations which were compromised by inaccurate delivery date predictions using current scheduling software. After the sale of the George Group to Accenture, Mike George Jr acquired an Aerospace and Defense company, Kessington Aerospace. One of the goals of this acquisition was to develop and prove-out scheduling software to accurately predict delivery dates.  The software development was enabled 1000 fold increase in compute power of the Cloud and the Internet of Things. Prior to the acquisition, Kessington had achieved less than 54% on time to promise date. The commercially available On-Time.ai software was developed, tested, and perfected, using Kessington as the test platform. The company now delivers >98% on time to promise date. You are welcome to visit the factory in Elkhart IN or by Video Conferencing for a live demonstration of the software using the contact link.

Kessington Plant

Multus Machining Department: These are 11 axis lathes with milling capability, SPC to detect and eliminate errors, and have 72 captive tools which reduce setup time.

Little's Law

Little’s Law* is often stated in terms of “Arrival Rates” to Work In Process. But since the queues of WIP are assumed stable, the Law is equally valid stated in terms of “Exit Rate” from Work In Process

*M. George et al “Lean Six Sigma in the Age of Artificial Intelligence” McGraw-Hill, 2019, pp. 10,55.

Figure 1

On-Time.ai Enablers in addition to the Internet of Things

Figure 2

Model for feedback to detect late jobs and correct job delivery to on time as an analogy to a ballistic missile solution 

Figure 3

Artificial Intelligence can look at all the jobs and find those which most nearly match the one currently being worked and provide a sequence which cuts setup time by more than 50% by solving the Traveling Salesman Problem.

Figure 4

Internet of Things: This hand-held bar code reader costs less than $100  (figure 4), is connected to the internet and eliminates human error in job and part number input by providing real-time data on the entry and exit time of WIP in each Pull Group. Reduction of capacity due to machine downtime and absenteeism is automatically incorporated in the corrective action feedback (figure 3) to ensure on-time delivery.  Our consultants will assist you in the installation of the Internet of Things.