This is the third installment in a conversation I recently had with one of our clients. (Some minor changes may be made in what is published below to protect the confidentiality of our client.)




As I said earlier, things may have changed dramatically in your workflows since I was last on-site with you. However, given what I knew then, here is how would envision production management and controls (from a high level) in a demand-driven environment using buffers to manage priorities and flows:

Mfg Workflow w Buffers 20151016.jpg

In this diagram, the hopper-shaped (trapezoidal) green, yellow and red shapes represent STOCK BUFFERS (actively synchronized replenishment strategic inventories). In this case, the one buffer symbolizes all of your raw materials inventories.

The semicircular green, yellow and red shapes represent TIME BUFFERS. Typically, a time buffer is created by taking the full time allotment for an activity, and dividing that time by one-third (at least as a starting point, remembering that buffers can and should be adjusted over time to optimize system Throughput). Then, that one-third of the full time is subdivided equally into three zones—a GREEN ZONE, a YELLOW ZONE, and a RED ZONE.

Allow me give you an example: Let’s imagine that you were going to create a TIME BUFFER at a certain control point in your system. And, for simplicity’s sake in our example, let us further say that this is a MTS (make to stock) environment and not MTO (make to order). In such an environment, there may be a whole class of goods for which it might be said, it should take 18 working-hours for a production order to get from the gating operation (release to production) to control point ‘A’.

We would take that 18 working-hours and divide it by three. That would give us a time buffer at control point ‘A’ of six hours (18 / 3 = 6). That six-hour time buffer would be subdivided into three equal zones: a two-hour GREEN ZONE, a two-hour YELLOW ZONE, and a two-hour RED ZONE.

The time at which a particular order should arrive at control point ‘A’ is calculated at the time it is released as being the release time plus 18 working-hours or, with more sophistication, as the next time-slot available in the work already queued up for control point ‘A’. In our example, and to make our math simple, let’s say that the calculated time for arrival of Order 1001 at control point ‘A’ is noon tomorrow. (Order 1001 is an order for a finished good that is scheduled to ship at a specified date and time.)

If Order 1001 arrives tomorrow morning (all preceding steps complete and in good order) between 6:00 AM and 8:00 AM, it will have arrived in the GREEN ZONE and be available to control point ‘A’. No action is necessary.

If Order 1001 has not arrived at control point ‘A’ by 8:00 AM, production managers should be aware (by some simple mechanism usually encapsulated in a visual system supported by a software application) that there is a HOLE in the YELLOW ZONE of the BUFFER for CONTROL POINT ‘A’; however, they would merely deem this as something of which to be aware, but not necessarily take any action.

If Order 1001 arrives at control point ‘A’ at 9:51 AM (between 8:00:01 AM and 10:00 AM), it would recorded as having arrived in the YELLOW ZONE. Life is still good!

However, if Order 1001 has not arrived at control point ‘A’ by any time after 10:00 AM (the beginning of the RED ZONE in this TIME BUFFER), managers should be ALERTED as to a HOLE in the RED ZONE of the BUFFER for CONTROL POINT ‘A’. This should trigger appropriate investigation, but not necessarily remedial action. For example, if a quick look tells managers that Order 1001 is within 10 minutes of wrapping up at the preceding work center or operation, they would likely need to take no action.

Nevertheless, whenever an order arrives at a control point in the RED ZONE or LATE, not only is the time of arrival recorded (as with all arrivals), but this should automatically trigger the mandatory selection for a REASON CODE (and, perhaps, some comments) to attach to the RED ZONE arrival record. This allows for the use of Pareto analysis of all threats to FLOW occurring in the system.

Priorities for action are easily determined and agreed upon


Such a system should present two pieces of vital information:

      1. Buffer Status of every RELEASED customer order at each selected CONTROL POINT
        1. Not due
        2. Arrived early (before the GREEN ZONE start) (datetime)
        3. GREEN ZONE
          1. A hole
          2. Arrived (datetime)
        4. YELLOW ZONE
          1. A hole
          2. Arrived (datetime)
        5. RED ZONE
          1. A hole
          2. Arrived (datetime)
        6. LATE
          1. A hole
          2. Arrived (datetime)
      2. Buffer Penetration – This is a simple percent calculation of the amount of buffer penetration divided by the total buffer size. For example, if an order has a six-hour TIME BUFFER (360 minutes) and it is 216 minutes late, its BUFFER PENETRATION IS 216/360 or 60 percent.


By comparing these two factors, priorities for actions can easily be determined using the following two, very simple, rules:

      1. RED ZONE or LATER (sometimes referred to as the BLACK ZONE) always gets worked on first; then YELLOW ZONE; and, finally, if any action need be take on GREEN ZONE items
      2. The HIGHER the BUFFER PENETRATION (percent), the HIGHER the PRIORITY for action and attention


An immediate comprehension of the ‘health’ of your system in a moment


By using STRATEGIC BUFFERS in this way, with relatively modest data maintenance, every executive and manager can tell in two minutes or less the ‘health’ of the system at any given time. It is simple:

      • IF BUFFERS are dominated by GREENS and YELLOWS, all is well
      • IF BUFFERS are showing significant number of REDS (or BLACKS), then something has gone wrong


No only so, but the PRIORITIES FOR ACTIONS will be very clear to everyone (as indicated above).


If you believe these concepts are worth pursuing, and you don’t feel that the [demand-driven scheduling system we suggested] will get the job done for you—for whatever reason—then there are other options that you do not have to build from scratch.

I would next suggest that you take a look at a product called [another demand-driven production control system we suggested].

I hope this explanation of how strategically placed and dynamically managed buffers can help you manage FLOW simply, yet effectively.




We love hearing from you. Leave a comment to share your thoughts and experiences related to the post.


If you would like to know the names of the truly demand-driven solutions we recommended to this client, drop us a request here.


Follow us on Twitter: @RKLeSolutions and @RDCushing

LIKE us on Facebook: RKL eSolutions and GeeWhiz2ROI