In this blog I talk about how to increase customer service while decreasing finished goods stocks by using the Glenday Sieve technique (also known as Green streaming).

The Glenday Sieve introduces the concept of a repeating production cycle to increase production efficiency. It can be used effectively together with automation to delay determining the production mix until the last possible moment.

The Glenday Sieve / Green Streaming concept recommends the manufacture the high-volume “A” items each cycle and the manufacture lower volume items “B” and “C”s less frequently. Items are manufactured in a fixed sequence e.g. always do product A then Product B then C, … etc.

To better explain the concepts involved (Glenday Sieve and Automation) I am using the example of a company in the plastics sector that extrudes different grades, micron thicknesses and colours of plastic film. The company then converts these into end products based on demand.

This combination of techniques can apply to any company that manufactures products for both stock and specific customer orders e.g. food, personal use items, pharmaceuticals, hardware, etc. These companies are described as either make-to-order or make-to-stock type companies.

In this example there are clear benefits to sequencing production to reduce setup times.

It is beneficial for the company to manufacture in a predetermined product sequence e.g. grade x in colour (light to dark), micron and width sequence, then grade y …, etc. to achieve these setup efficiencies.

The question to be answered is what products to make in each production cycle.

These techniques can be combined and implemented in the following way:

  1. Cycle Length
    • Determine a production cycle length e.g. the factory must be able to cycle through all products in the chosen cycle period (a week / fortnight) and meet the S&OP volume requirement. If the cycle period is too short then setups will consume too much production capacity, if too long then finished goods holding will rise and customer service levels may suffer.
      • There are many trade-offs to be made here e.g.
        • adding shifts to increase the capacity in a cycle.
        • Running lower demand items every second cycle to reduce the setup time requirements.
  2.  Allocating production orders.
    • We now have a production cycle length, associated available capacity and theoretical product sequence. All that is required is to replace the theoretical sequence with actual production orders.
      • This should be done programmatically to reduce the time and effort needed to complete the task.
    • This can be done by prioritising the loading of products in a priority sequence until the production capacity of the cycle is consumed e.g.
      • Load actual customer orders due in an agreed time horizon (e.g. due within 2 cycles), first.
      • Then load “A” items required to meet finished goods stock requirements.
      • Then load make-to-stock items where actual customer orders exist. There will be backlog orders for these items.
      • Then load “B” items based on the priority of days to run-out i.e. most vulnerable to run-out, first.
      • Load the remaining items in priority sequence until all planned capacity is consumed.
      • Adjust volumes of batches if the capacity is not fully consumed or over-consumed.
      • Note – items will only be loaded into the sequence if all supporting input materials are available.
    • The allocation sequence can be repeated daily for the remainder of the cycle. An example might explain this better.
      • Assume a weekly production cycle.
      • On Friday, fix the cycle contents for the next week’s production cycle.
      • On the next working day
        • update the schedule with actual production results and reschedule.
        • freeze the schedule for the next 2 – 3 shifts, so as not to impact on paperwork already released to the floor.
        • and reload the sequence, see 2 above, for the remainder of the week.
      • Doing this will immediately
        • take any new customer orders or backlog items into account
        • Remove the lowest priority items from the sequence if production delays have been experienced
        • Add items to the schedule if production is running more efficiently than expected.
    • This approach makes the plant very flexible concerning the production mix. The plant will react to actual mix of sales at the beginning of each cycle.
    • This process can be repeated for as many weeks as the company needs to create a detailed schedule. This may be necessary to determine the timing for supporting raw materials.

The benefits of using this approach are

  1. The bulk of the scheduling task is automated. The Scheduler will still be needed to
    • Set production cycle durations and capacities.
    • Set production sequences within a cycle.
    • Deal with exceptions e.g. non-availability of input materials
    • ETC.
  2. It may not be necessary to run MRP. This depends mainly on the number of and lead times of input materials
  3. The system will react almost instantly to backlog orders while ensuring that A items are kept in stock, resulting in an immediate improvement in customer service levels.
  4. The system can be adapted for online order entry i.e. the customer can enter their order directly into the system and get a valid delivery date promise immediately.

Could you use this scheduling approach in your factory?