Whether implemented as a stand-alone solution or integrated part of the ERP or MRP system, APS can help manufacturers adapt to their internal constraints and changing customer requirements, while maintaining high profit margins.

By Sami Cassis

With such a strong endorsement and the availability of several mature and extended functionality packages on the market, one would expect rapid and outstanding benefits each time an APS system is implemented. However, the fact is that although some implementations have been a resounding success, others have gone over some rough roads before yielding the expected benefits. For this reason, it is important to question what are the elements of success, and what are the pitfalls to avoid in implementing an APS system?


The Nature of Advanced Planning Systems
Before we examine success and risk factors, let us briefly review the nature of advanced planning systems and how they differ from conventional MRP/MRP II or ERP systems. At first glance, APS systems appear to function just like any other manufacturing system. They use routings and bills of material; and they need information about inventory levels, orders and so on. However, there are subtle differences between APS and ERP systems. For one, APS systems do not require — and should not ask for — "queue" time or "lead" time. Instead, they require specific information per operation, mostly the production rate and the resource(s) or constraints. This seemingly minor difference, if properly exploited, could have a major impact on the implementation of APS systems, as we will see later on.

1. Project definition and scoping •Problem definition and setting project objectives •Environment analysis and defining the project scope •Determining which problems will be addressed through BPR instead of IT •Establishing initial time frame for the project, including each individual activity 2. Model development and refinement •Model development and refinement until the system optimizes the behavior of the production environment 3. Integration and data transfer with the existing legacy systems 4. Parallel execution, users training and education

Using that data, APS systems attempt to minimize queue time and, hence, maximize material flow within a factory. Although some systems accomplish that goal better or worse than others, the quality and level of optimization of the system is rarely a major element in the success, or lack thereof, in an APS implementation. One should remember that in most, if not all, cases the plan produced by an intelligent APS system will be superior to a plan produced manually. Discarding the "optimization level" as a critical success element does not, however, mean ignoring all features of the software. There are some basic features and functionality which will contribute significantly to the success of an APS system.


Such as:
•The ability to simultaneously constrain material and capacity. In some manufacturing environments, material might not be a constraint. In these cases, a capacity-based APS system will do the job. However, if material must be a constraint, then attempting to use two systems — an MRP system for material and an APS system for capacity — will not work, due to the tight interaction between these two constraints. Having one system managing both resources — material and capacity — will lead to more integrated planning, less iterations and a better overall optimization.

•The ability to produce an "intelligent" plan that does a good job of synchronizing operations and sub-assemblies, while delivering on time. Most intelligent systems today offer this kind of synchronization feature. The days of comparing "forward" to "backward" systems are over. The jury has reached a verdict: They are both "ugly." Forward systems generate too much work in process (WIP), while backward systems are too much Just-in-Time (JIT) and cannot properly manage safety buffers. Today's intelligent systems are all multi-passes with constraints optimization logic.

•In the case of a multi-plant environment, the ability to accommodate multiple factories under the control of one or several planners becomes critical. Such environments can be complex, especially if the plants feed each other, and hence must be synchronized.

•Finally, it is important to ensure that the APS system can model the "critical constraints" of your environment. Companies shouldn't worry about a detailed model of every operation. APS systems can model an environment to a degree of precision far exceeding the data accuracy of most enterprises or their reliability of execution. The features set of a modern APS system could far exceed the length of this article, when you take into account operations overlap, alternate routings, batching, tooling families, lot splitting, etc.

The only feature that really counts is the ability of the system to model your main constraints. For example, if your bottleneck is an oven, make sure the system can batch several jobs in one run. In short, don't worry about non-constraints. Avoid the micro-modeling syndrome. It will only complicate the project with no added benefits.

Armed with a system whose feature sets match the critical requirements of your environment, you are now ready to face the implementation phase of your project. We will not spend time going through the generic elements of success required on any project — e.g., executive support and commitment, and having the necessary resources. Rather, we will focus on factors specific to the success of APS implementations.


User Support: A Critical Element Of Success
Getting commitment from company executives, support from the implementation team and enthusiasm from the users are critical success factors. APS systems are as difficult to implement as any other manufacturing systems. The model must be developed, data must be transferred between systems, and people must be trained to use the system effectively. At the final stage of the implementation, the system will demand more from the implementation team before delivering value. In these moments, you will need all the support and commitment possible. And that level of commitment cannot be attained unless users are as convinced and passionate about the project as yourself.

It is important to take the time to explain the impact the system will have on their lives, as well as the benefits it will bring to them and the corporation. APS systems can be difficult to understand and conceptualize. Concepts such as material buffers, release management, and operation prioritization based on overall job synchronization are quite complex. Aside from the issue of complexity, these systems, in most cases, are very different from established planning and scheduling processes, and often demand major changes to the business environment.


Countering Resistance To Change
The best way to fight resistance to change is to make the users themselves the catalyst of that change. This can only be accomplished if they have full ownership and empowerment within the change process. The software provider should focus its energy on helping users understand the concepts and dynamics of their environment, and allow them to reach their own conclusions and decisions for change. Then they will have what Dr. Eli Goldratt refers to as the "passion of the inventor." When users feel that the ideas are their own, you will have achieved proper support for the project.

In fact, I have seen several successful implementations of what I would have defined as "totally inadequate software tools" due to the ownership the users took of the project. Today, the most personal, popular and easy-to-understand planning and scheduling tool on the market is not made by a manufacturing software company but by Microsoft. I am referring to the Excel spreadsheet. I have seen more plants scheduled with Excel than any other tool. Should a deadly virus find its way to Excel, the whole U.S. manufacturing economy would probably grind to a halt. And yet, in reality, the planning and scheduling capabilities of Excel are far below the functional features of the smallest planning and scheduling software package. The difference is that with Excel, users feel in control of their environment and can make their own decisions.


Focus On The Key Elements
One of the most annoying elements of any manufacturing project is the ocean of data that must be gathered for the system to operate properly. Part number, resupply lead time, queue time, rate, yield, etc., are only part of the data requirements of manufacturing systems. The issue is compounded by the fact that this information is required for every part and operation. This overwhelming data requirement stems from the lack of intelligence in most manufacturing systems and their inability to discriminate between critical and non-critical information.

Nothing is more frustrating than having to specify lead times by part and operation. A user once told me that he feels captive to his lead times.

Question: If the resulting schedule is not accurate, who is to blame? Answer: the user who entered the lead times.

It is as if the system asked you for the solution to the problem, and if the results do not match the reality of the shop floor, then you are to blame. No wonder so many manufacturing systems do not make a dent on the shop floor.

On the other hand, an intelligent schedule for a manufacturing facility can be easily generated by focusing on the main constraints of the environment. By specifying the manufacturing parameters (rates and set-ups) for these constraints, a plan or schedule can be obtained. Most of today's intelligent systems can achieve that objective. The relief — in gathering and entering data — provided by the system can then be used to improve the accuracy of the data provided for the main constraints. This, in turn, drastically improves the quality of the schedule provided by the system.


Solutions To The "Loss-Of-Control" Syndrome
A planner's biggest concern during the implementation of an APS system is the potential loss of control. People do not feel comfortable entering data in a black box, running an obscure optimization algorithm, and getting results that are not intuitively what they would have done. Even if the resulting schedule is better than the manual one, the planner will not trust it.

To compound the problem, a computer system cannot take into consideration all the parameters that the planner needs. How do you factor in the fact that ORDER-123 must run Wednesday at 10 a.m. because the customer will then be at the plant to approve the color rendition? Have you tried to do that with an APS system? How do you explain to a system that you can delay one of the two orders for a specific customer, but not both?

What the system needs to do is work jointly with the planner. A initial plan must be set by the system. The planner, however, should have the right to edit that plan. Any edit should be preserved through multiple runs of the engine and can only be released by the planner himself.


Walking The Line Between Environment Modeling And BPR
One customer in the remanufacturing business had an automated conveyor on which products would turn endlessly until they were randomly picked by an operator. Since the selection process was purely random, the cycle time of an order was about as predictable as the numbers in the national lottery.

Modeling that environment as it existed would have been pure folly. In attempting to do so, we would have engraved that randomness in the system itself, sealing the fate of our cycle time reduction program. An initial, rather crude approach was to use colored hangers that would reflect the time when the product was put on the conveyor. By picking the earliest products first, cycle time was drastically reduced. I have encountered several similar cases where there was more value in reengineering the process before modeling it. It is important to learn when to rely on a system — and when to rely on common sense — to solve a problem.


Is It Worth the Effort?
By evaluating their situation realistically, following the elements of success and avoiding the obvious — and not so obvious — pitfalls, most manufacturers can expect to reap significant benefits from an advanced planning system, whether it is implemented as a stand-alone solution or as an integrated part of the ERP or MRP system.

Properly modeled and implemented, the APS can help manufacturers adapt to their internal constraints and changing customer requirements, while maintaining high profit margins. Typically, the results realized include improved manufacturing capabilities, reduced costs, increased productivity and enhanced customer service. In fact, for many manufacturers, the advanced planning system is the new strategic weapon of choice for competing into the next century.

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]