On-time delivery (OTD), a measurement of the level of finish goods or services delivered when promised and in full, is a signal of efficiency that reveals how well operations is doing. This KPI can be significantly improved, even when it appears very challenging—because of, say, new wrinkles in manufacturing, such as mass customization.
Exxelia Dearborn, an electronic capacitors manufacturer specializing in film capacitors and EMI/RFI filters, steadily obtained an OTD rate of 79% with a high-success mark of 82%. Multiple efforts and teams were tried and implemented to do better. But the typical excuses and obstacles were presented time and again for why the OTD rate was not improving or could not get above 82%.
However, after receiving critical aerospace and DOD customer feedback, Dearborn managers got a bit OCD (in a healthy way): They set the goal of reaching 95% OTD—and then the company actually reached 99% OTD and consistently stayed there.
To improve OTD, start by looking at the problem areas. Manufacturers know that OTD is affected by a multitude of factors, such as shop floor scheduling, yield estimates, ill-informed promises, generic date ranges, inventory errors, inexperienced planners, shipping mistakes, spec errors and vendor rejects. Each factor contributes to decreasing the OTD rate, and it does not take a major problem with one area to significantly impact OTD. For Dearborn, three areas were determined and recognized as the most critical factors:
This problem is a combination of ill-informed promises and generic date ranges. When a customer finalizes an order, employees at many manufacturers still guesstimate the production time and delivery date because they do not know an accurate production time and do not have an accurate analysis of delivery history.
This issue centered on scheduling and priority of production runs. Dearborn averaged around 2,000 manufacturing orders, making it difficult to easily determine which order to do first or which order needed to move up in order of priority. The result was the shop floor often worked on an incorrect order in terms of priority.
This is a common problem: Plant managers, planners and production managers know that in manufacturing a product there will be a percentage of products rejected or lost. Manufacturing in some industries can lose 45% during a production run.
The question to ask is: Do your employees have a tool to help them accurately determine how much to start with in order to successfully produce the desired quantity?
Manufacturers also need to realize that the problem with OTD is not due to the various programs and systems, such as ERP, MES and inventory. The root problem is in the interpretation of the data from those systems. Humans are being asked to make exact decisions based on scattered, disparate data and not from comprehensive integration and analysis.
Lastly, present the right data to the right person for the right decision.
For Dearborn, a tool was provided. This tool is a decision-making platform that connects the various systems, as well as integrates the data to produce analyzed information. The platform drills into the systems for the appropriate data, routes relevant data via dashboards to appropriate personnel and incorporates artificial intelligence to automate simple decisions.
When a complex decision is needed or not enough information is available, the platform presents to an internal expert all relevant data and information so an informed decision can be made based on that expert’s experience and insights.
For Dearborn, the decision-making platform became magic that so far has translated into an extraordinarily good OTD performance rate for more than 40,000 manufactured parts. Burdensome responsibilities and inefficiencies were removed, and guesswork was replaced by data, specifics, accuracy and precision.