ON TIME DELIVERY: In job-based production, it may not be what you think

By: Kent Vincent

On time Performance

One of the only metrics that attempts to look at time as part of overall performance monitoring is the on-time delivery (O-T-D) measure, which attempts to report the percentage of jobs that reach their destination (internal or external) on time when compared to the planned arrival time. There are serious drawbacks to this indicator, however.

While on-time delivery is desirable as an outcome the emphasis on it as a performance measure can border on the dysfunctional.  Instead of helping to reduce the end-to-end time span through the entire process chain, internal departments and external suppliers both tend to pad their planned or promised lead times so that on-time-delivery performance looks good.

To start with, a perverse incentive is often put in place to encourage a manager to inflate planned lead teams, thus institutionalizing a longer total end-to-end production and delivery time (what we call a “critical time” that is described in other posts).  Second, it can encourage managers to start jobs ahead of time when openings appear in a production schedule, causing in-process production to sit idle and add to work in process costs and total time consumed. Third, when jobs that were not yet needed were started, it could supplant urgent production and create a ripple effect of unintentional negative issues:

  • The job that was already started might have had a long setup time, and workers would be reluctant to tear it down so that urgent jobs would be delayed. Or the workers are forced to tear down the present setup and work on the urgent job, but this would mean that the setup would have to be restaged and repeated, thus adding extra setup to the workload. Also, the job that was finished early may take up floor or dock space and moves may be required – non value added activity that may introduce handling damage, misdirection or other consequences.

Similar effects occur with the supplier base: Knowing that they are being measured by O-T-D metrics, they try to quote the longest lead time that they can get approved.  This is especially true when the supplier sees that the customer’s forecasts are habitually unreliable and the customer changes order quantities often. Then the padded lead time gives the supplier additional wiggle room to cope with these changes.  The same phenomenon can play out in the back office support functions such as quote generation, order assembly management, and related processes and procedures.

In many cases things can degrade further. Individual departments padding planned lead times actually cause worse performance overall. The domino effect leads to excessively long lead times for both components and finished goods.  This leads to a host of other problems that ultimately translate into poor order fulfillment. Ultimately, production processes and procedures are likely to deteriorate rather than improve.

Authors such as Schonberger (Lean Six Sigma Best Practices and Let’s Fix It!) point out that On-Time-Delivery and Operational Equipment Effectiveness (OEE) are derived or spin-off metrics arising from more basic measures of quality and efficiency that deserve more attention.  Specifically, On-Time-Delivery is really a derivative of the “golden goals” of Quality, Responsiveness, Flexibility, and Value.(below) Pay heed to the first four and On-Time-Delivery through application of lean and total-quality tools, and the on-time performance follows fairly reliably.  But there are few, if any standardized, established set of tools that direct themselves squarely at on-time performance, at least none that preclude “gaming” the system with lead time and inventory padding as described above.

Primary Golden Goals

Aside from focusing on the primary performance goals as a route to on-time delivery, the root time element to attack is total “critical” time throughout both the procurement-transformation-delivery cycle.  This macro level time span is much like the critical path length concept as applied to project structure where the linked constraining segments of time are monitored in preference to time segments that feed into the main pathway or await completion of critical tasks. . For this metric all component times should represent delivery of resources from production or supplier sources rather than from stock. So, for instance, if a housing can be pulled from stock in one day, but it takes 8 days to replenish it via off site production, then the larger figure should be used.

If we pair this idea up with that of a Gantt chart we have the makings of a Critical Time map, alternately known as a Manufacturing Critical Time map or a Fulfillment Critical Time map depending on whether we’re dealing with manufacturing or distribution.

In the example below the as CT map shows processes involving sourcing of component parts and preparation of a housing occurring in parallel, then leading to final kitting and assembly.  The blue areas are key physical value-added steps bounded by proportionately large margins of “white space” that represent idle or queue time plus any incidental transfer and conveyance.  It is this white space that represents the best and easiest opportunity to compress overall time leading to a naturally improved on-time-delivery metric without “gaming” the system or just shifting the goal posts.  In the lower diagram the white space time has been excised by cellularizing the kitting operations and by employing various mechanisms to pull product along before it can stall in these queues and batches for lengthy periods. Some additional improvement was no doubt achieved by revisiting the sourcing options for key components and the length of resupply time.  Also note how easily identifiable and visible the time reduction opportunities are compared the way they may be depicted in other diagrammatic methods such as value stream and swim lane maps to name a couple.  And these reduced times add up to not only faster delivery time but in a great deal of freed up cash not hung up in work in process queues or inventory stores.  Other elements of supply chain fulfillment time can then be addressed such as transit, routing optimization, and so on.


Bottom line: strive for on-time-delivery improvement as an inevitable outcome of improvement in the key critical elements of the end-to-end source-produce-deliver rather than as a metric to game or defeat, and the time and the cost reductions will follow, with any luck, better than they ever have before.

About the author

Kent Vincent works with small to midsize manufacturing, logistics companies to make things operate better, faster and cheaper, and employ a variety of methodologies in the process. He is particularly adept at helping high variety low-volume businesses and companies that operate on a “vat-level scale”. He has been recognized and awarded for improving material work order flow, waste reduction, process improvement, and process re-engineering.

For More information:

Industrial Systems & Re-engineering Services
6607 Fernwood Drive
Lisle, IL 60532
Email: Isandr@usa.com

Take action today and enjoy benefits tomorrow – contact us!


100 High Grove Blvd,
Glendale Heights, IL 60139

What's On Your Mind?


“Our company was recently purchased and the new management wanted our products to be manufactured in China. We have a staff of two and the thought of dealing with people on the other side of the planet, in a different time zone, a different language and a different measurement and monetary system was giving me …

—MG in Milwaukee, VP of Finance for a tier two automotive products manufacturer.

Follow Us!