MRP II SYSTEMS have been broadly accepted due to the extensive and useful functionality they provide, even in those situations where the planning/decision support paradigm provided by the system poorly fits the business.

By Philip Rhodes, CFPIM

Implicit in the planning/decision support paradigm of MRP II systems are the following hidden assumptions:

• materials planning is of primary importance
• capacity planning and scheduling are separate and secondary functions to materials planning
• the production of individual items (finished goods and components) can be planned (in production and for distribution) independently of other "related" items
• manual evaluations and reviews of system-generated materials plans (including the master schedule) and extensive manual changes to those plans are practical
• multi-plant issues and distribution economics can be handled outside of the system.

Today's MRP II systems are direct descendants of systems which go back well over 30 years. These early ideas, developed for fabrication and assembly manufacturers, have been highly elaborated and made more flexible with hundreds, even thousands of imbedded application parameters and logic tables. The result of all this is an appearance of flexibility which can be misleading. Let's examine the MRP II paradigm in greater detail.

A SIMPLE JOINT REPLENISHMENT STRATEGY Equal Periods of Supply is a basic joint replenishment strategy which, as with all joint replenishment strategies, is not supported by MRP II. The vertical axis represents periods of supply. The width of each column represents the rate of demand for an item in the family being jointly planned. This could be the same packaged item distributed to three different warehouses or the same product packaged into three sizes. The pitcher represents one batch of product being allocated across warehouses or package sizes. The dashed line is the level to which each column is to be filled.

Examining the MRP II Paradigm
Under the MRP II paradigm, a plantwide materials plan is driven by an updated master schedule. The master schedule (MS) itself is updated periodically (e.g., weekly) to extend the plan further out in time and to reflect newly booked sales orders, updated sales forecasts and recent production.


Job Shops and Capacity Dominated Production
The MRP II planning paradigm is irrelevant to job shop production. Job shops typically integrate each new job (produced to the customer's drawings/specifications), one by one, into an existing plan. There is no periodic update of a master schedule and seldom a need to run MRP.

This is because job shops issue purchase orders on a job-by-job basis, synchronizing receipts with planned start dates. (Note: Some commonly used components, such as bar stock, may be carried in inventory with replenishments triggered by order points.)

The MRP II planning paradigm is also clearly inappropriate where capacity is of overwhelming importance and material availability is a very secondary concern. Rolling mills, semiconductor plants and many chemical process plants are examples of this.


The "Item Independence" Assumption
A critical limitation of the MRP II planning paradigm is that it develops a materials plan for each item independently of similar items.

In many production situations, "families" of products must be produced together — and often in a particular sequence — in order to minimize production costs. In this situation, materials plans which use standard production lead times in planning have little validity.

For example, if the items in a family are produced together (in sequence) every fourth week, the production lead time for a particular item could be one day at one point in time and four weeks at another. It all depends on where you are in the family's production "cycle." Assuming an average lead time (e.g., two weeks) for planning purposes leads to meaningless plans.

A second common situation where the "item independence" assumption is inappropriate for planning purposes is the following. Many products (pharmaceuticals, lubricants, paints, etc.), for a variety of reasons, are produced in a single batch size and must be immediately packed. They cannot be inventoried in a bulk state. In addition, it is often undesirable to pack an entire batch of the bulk product into a single package size.

Because MRP II plans each package size of the product independently of its other sizes, any MRP-generated materials plans for the individual package sizes are, at best, only a starting point for manually developed plans.

When multiple items need to be planned together as in the above examples, this is called "joint replenishment." Joint replenishment issues also arise frequently in purchasing (invoice level discounts) and distribution (e.g., the need to ship in truckloads).

In order to effectively address joint replenishment economics, it is often necessary to integrate planning with scheduling/execution issues. One cannot deal with one issue and then, in separate steps, deal with the other. Yet, this is exactly the way MRP II systems work.


The DRP Limitations of MRP II
Although distribution requirements planning (DRP) extends the MRP II concept of time-phased materials planning to distribution, it continues to ignore joint replenishment issues and does not integrate distribution planning with master scheduling from a decision-making viewpoint.

DRP is not an execution system and does not support the many critical and complex decisions made every day by distribution and traffic personnel. For example, DRP systems typically:

• do not support automation of the process of intelligently increasing/decreasing shipments to distribution warehouses to meet truckload requirements.
• do not identify inventory imbalances between satellite warehouses and consider transshipments between them as alternatives to replenishments from the master warehouse.
• do not recognize when to "ration" warehouse replenishment orders to avoid premature stockouts at the plant or master warehouse. This inadequacy leads to unnecessary customer backorders and costly revisions to master schedules.
• do not provide for "push" concepts in replenishing satellite warehouses. Under a push approach, most or much of finished goods production is immediately staged in shipping where it is pushed to satellites using an equal periods of supply concept. This strategy avoids handling costs at the plant and places the inventory at the satellites where it is immediately available to fill customer orders.

The Master Scheduling Limitations of MRP II
The master production schedule (MPS) drives MRP. The master scheduling module in MRP II systems typically is used to generate a materials plan for finished goods (in an assemble-to-order environment the master schedule is for "highest level" stocked assemblies).

The resulting plan is then manually manipulated, hopefully considering customer promise dates and backorders, the availability of production resources, the costs of flexing those resources (e.g., overtime and sub-contracting) and opportunities for joint production economics. Placing this burden on the personnel responsible for master scheduling is a very tall order.

To support this complex process, MRP II systems offer only rough cut capacity planning and infinite capacity planning — after completion of the MRP run — to assist with these complex decisions.

To be specific, these systems typically fall short in the following respects:

• Where more than one plant is capable of producing the same item, they offer no guidance on the allocation of production among plants.
• They offer no guidance on the possible sequencing in production of items, which use the same or similar setups, to minimize production costs (a joint replenishment situation).
• They offer little guidance on "opportunistic production," taking advantage of the need for one item to produce other, related items.
• In overload situations they offer no guidance beyond pegged requirements on where and how to reduce production quantities.
• They do not provide facilities to automatically generate alternative master schedules or provide any overall measures of "goodness" (beyond total inventory dollars) with which to compare one master schedule against another.

A COLLECTION OF DEPARTMENT SYSTEMS Each of the department systems of MRP II, while drawing on an integrated database, focuses on a narrow view of the total picture. The result is the need for extensive manual review and intervention.

The Systems Integration of MRP II Systems
Many people, particularly those from the information systems side of the business, confuse systems integration with integrated decision-making.

A critical look at MRP II systems reveals that these systems, while providing systems integration, do not provide integrated decision-making. In fact, MRP II systems are essentially a collection of tightly system-integrated departmental systems. They are systems which deal sequentially with one aspect of the planning and scheduling problem at a time.

It is that super human, the master scheduler, who is expected to pull it all together. He/she is the one who is expected to implicitly evaluate trade-offs, modifying plans to arrive at the best balance of conflicting objectives, and do this in a tight time frame.


Selecting MRP II and Other Software
Companies planning the purchase of an MRP II system need to determine how well their production and distribution economics fit the MRP II paradigm.

A poor fit is an alert that the core master scheduling and planning functionality of MRP II systems may be more difficult to implement and less effective than anticipated.

The system architectures of some MRP II systems reflect the attitude that it is possible to be all things to all people. Some even try to dictate "best" practices. This results in systems which are not only highly complex, but very difficult to modify or interface with specialized decision support systems.

System buyers need to be inoculated with a healthy dose of skepticism concerning how closely a package can be made to fit their decision support needs without extensive modification or interfacing to specialized software products.

Once top management has been sold by software sales people on the "flexibility" of the MRP II system being pitched, they have a hard time understanding why the implementation is so drawn out and costly, and why benefits are so much lower than anticipated.

Should it become apparent that additional software is required to supplement an integrated manufacturing software product, it is important to determine the "openness" of that software product to integration with other systems.


The Quality of System Openness
Openness in a software product is, in equal parts, a matter of technology and application design. A seemingly open technology can become closed by an overly controls-minded software product, or by a product which uses standard technologies in a non-standard manner.

This quality of openness is claimed by many manufacturing systems, but a claim and reality are not the same. Perhaps the simplest and best way to evaluate openness is to talk with other buyers of the system being evaluated. Find out from them what it took to modify the system and how long to payback. And don't just speak to those who had responsibility for selecting the system. Speak to those in the trenches who had to make the system pay off.

Copyright © 2020 by APICS — The Educational Society for Resource Management. All rights reserved.

Web Site © Copyright 2020 by Lionheart Publishing, Inc.
All rights reserved.


Lionheart Publishing, Inc.
2555 Cumberland Parkway, Suite 299, Atlanta, GA 30339 USA
Phone: +44 23 8110 3411 | br> E-mail:
Web: www.lionheartpub.com


Web Design by Premier Web Designs
E-mail: [email protected]