Click here to set up your 60-day trial period of Prestige Scheduler.

Pivotal Z LLC.


Buy your own copy of Bill Kirchmier's new book below.

Finite Capacity Scheduling

Requirements For Creating Feasible Schedules

“Behold the turtle. He only makes progress when he sticks his neck out.”
— James B. Conant


Material Requirements Planning (MRP) systems claim to have both planning and scheduling. However, MRP with its infinite capacity backward pass (ICBP) approach is not capable of producing schedules that can be executed on the shop floor. This is a major distinction that needs to be acknowledged. Infinite scheduling continues to be used by MRP systems offering planning results but supplying limited or no scheduling functionality. Its long-term use has created a feeling that infinite scheduling is an acceptable and effective scheduling method. The best that can be said about infinite scheduling is that it functions only as a rough cut planning system. Much has been published about the inadequacy of infinite capacity scheduling.

Current planning technology is presented as the process of calculating material and capacity based on forecast orders, actual orders, or both over a specified time. The terms forecasting and planning both imply error. The primary source of the error is that both are based on periods, or buckets, of time. The output document submitted to the shop floor from an infinite capacity system is a dispatch list that specifies which products are to be produced and in what quantities during each bucket. The document will indicate what work should be done during the period of time, but does not offer the exact sequence of the events that will take place at each work center. This dispatch list typically facilitates a manual process where the manager determines the schedule and identifies the sequence of tasks going through each work center. This approach limits the ability to coordinate work across multiple work centers throughout the entire plant thereby restricting operational excellence. Infinite scheduling is a 30-piece band with each musician playing a different tune.

In contrast, scheduling implies accuracy. Scheduling is the process of accurately defining the operating conditions for production on a minute-by-minute basis through a bucket-less view of the operations timeline. The scheduling function covers a relatively short interval of time when compared to the planning cycle; the scheduling is tactical whereas planning is strategic. Scheduling as a tactical tool means that any resource that will affect production needs to be accurately scheduled (i.e., machines, personnel, fixtures, tools, outside vendors, etc.). The output document, commonly referred to as the work center to-do or work-to report, generated by a finite capacity scheduling system, accurately defines the sequence of all tasks through each work center on a minute-by-minute (or an even smaller unit of time if justified by the application) basis and coordinates all work centers to improve overall plant throughput.

The fact that FCS schedules are calculated to such accuracy causes concern that it may not be possible to follow this schedule. With valid data input and the discipline to follow the to-do lists, this is a baseless criticism. If the schedule is not calculated to this degree of accuracy, errors would be created in the schedule before it gets to the shop floor. This degree of accuracy is essential in order to optimize execution on the shop floor.

Actual conditions will oscillate around the FCS-scheduled conditions but the standards used to formulate the schedule establish acceptable limits of performance. When the actual conditions deviate to an unacceptable degree, such as a resource failure that requires being removed from production for repair, an update of conditions is required and a new schedule is created. When routines are reasonably accurate and data input is reliable, following the schedule will never be a problem, and schedules are often followed to the end of the expected cycle without the need for interim rescheduling under normal operating conditions.

The chart illustrates the risk of relying on infinite planning. The period-to-period capacity of the manufacturer is 1600 shop hours. The infinite loading overbooks the plant throughout most of the 13 week planning period. The results are excessive expediting to meet unrealistic due dates. Finite Capacity Scheduling fits the work load within the actual capacity, and also considers all constraints in the production environment such as tooling and manpower.
Consider the results of the Finite vs. Infinite Loading graphic in Figure 1.0. The Finite example is the classic reality presented by an actual production application. Study the graphics and compare finite loading reality to the impossible infinite loading example. Impossible because the infinite loading example consumes more shop hours capacity than is actually available across most of the period. The light shaded finite example is consistently below the actual total capacity available in the center plate and is feasible. It is not possible to consume more than the total available capacity, yet the majority of companies in the manufacturing sector continue to apply infinite capacity technology. Return often to this graphic as you progress through the book and consider the implications.

The chart above illustrates the risk of relying on infinite planning. The period-to-period capacity of this manufacturer is 1600 shop hours. The infinite loading overbooks the plant throughout most of the 13 week planning period. The results are excessive expediting to meet unrealistic due dates. Finite Capacity Scheduling fits the work load within the actual capacity, and also considers all other constraints in the production environment such as tooling and manpower.

In summary, infinite planning is a crude approximation of future conditions over an extended period; scheduling is an accurate list of events based on current conditions over a shorter interval. Some industries may be able to continue to function and remain competitive with an ICBP planning system without the need for a detailed scheduling system, but they will be the exception.


A variety of production examples exist in manufacturing companies with flow shops and job shops at opposite ends of the spectrum. Although FCS is a benefit to companies throughout the spectrum, the benefit tends to be more prominent and necessary for the job shop end of the spectrum. Most companies operate somewhere between these two extremes. To be competitive in the future, companies will have to resort to better scheduling technology. Planning systems alone will not fill the bill.

One of the author’s was recently working at one of the major Contract Manufacturers (CMs) for high tech equipment. This company has over 50 factories throughout the world and they build the equipment that the Cisco’s, Dells, and HPs of the world sell. They currently produce 90 percent of the inventory to stock (BTS – Build to Stock) and about 10 percent to order (BTO – Build to Order). They estimate that by 2015, it will shift to about 60 percent BTO and 40 percent BTS. This dramatic shift demonstrates a reorientation towards the job shop end of the spectrum, even for organizations that have traditionally been highly repetitive in their manufacturing processes.

Figure 1.1 The Management Spectrum shows how material and capacity management varies as product variety moves from High Volume Low Mix (HVLM) to High Mix Low Volume (HMLV). Historically the predominant manufacturing production model focused on simplistic HVLM products. The HVLM configuration is typically a linear (cellular) flow production model where material is the primary condition to manage, and capacity is the secondary condition to manage. At the extreme HMLV end of the production spectrum, the importance of materials management approaches zero when compared to the importance of capacity management requirements.

At this point capacity becomes the primary condition to manage, and material is the secondary condition to manage. Historically, the high volume model has been the model of choice for most manufacturing companies. Recently customer demand for more product diversity has increased the number of companies driven away from the High Volume Low Mix (HVLM) production model.

The success of HMLV production model depends on the ability to accurately, effectively and competitively schedule a high variety of products in small unit quantities; or any other combination of order mix that causes scheduling complexity.

Competition for resources at the operation level is the primary cause of scheduling complexity; excessive complexity leads to chaos. Managing this complexity requires a system that is capable of changing chaos to order, and FCS meets this goal.

Figures 1.1 and 1.2 are used to describe the distinction between HVLM simplistic flow models and highly diverse HMLV routines that deliver feasible schedules when the demand is complex routines requiring advanced scheduling technologies.


Figure 1.1 – Scheduling Spectrum,


Figure 1.2 – The Management Spectrum

The following excerpt by John Sprovieri, Senior Editor of Assembly Magazine’s article titled Managing High-Mix Low-Volume Assembly is an excellent description of industry direction.

“In describing U.S. consumers on his 1972 comedy recording, Class Clown, George Carlin was almost prophetic about the state of U.S. manufacturing today. The typical consumer, he noted, does not merely want an inhaler with a nasal decongestant. He wants an inhaler for his left nostril imprinted with his state’s motto.

Carlin may have been riffing on crass commercialism at the time, but his gibe goes to the heart of an issue facing a growing number of assembly plants today: Manufacturing is shifting away from high-volume, low-mix production to high-mix, low-volume production.

According to ASSEMBLY’s annual Capital Equipment Spending Survey, there’s little doubt that fewer manufacturers are performing high-volume assembly today than there were 8 years ago. In 1996, 25 percent of plants produced more than 1 million assemblies annually. In 2004, only 19 percent did so. At the same time, the number of low-volume manufacturers is increasing. In 1996, 24 percent of plants produced less than 1,000 assemblies annually. In 2004, 35 percent did so.

As production volumes have decreased, production variety has increased. U.S manufacturers are assembling a greater mix of products than ever before. In 1996, 16 percent of plants assembled more than 10 different product types annually. In 2004, 23percent did so.
The main reason for this shift, as Carlin alluded to, is that consumers increasingly want products tailored to their particular needs. Consider the humble refrigerator. It has one job: keep food cold. Yet Whirlpool Corp. (Benton Harbor, MI) offers no less than 57models of refrigerator, differentiated by configuration, features, finish, height, width and, of course, price.”

The entire article “Managing High-Mix, Low-Volume Assembly” by John Sprovieri can be found in your search engine.


Lean technology has received and continues to receive a high level of attention. Lean is a change management methodology adapted from the Toyota Production System (TPS) techniques. The focus of lean is “waste reduction.” Waste can be found in capacity, materials, labor, and many other resources within an enterprise. Lean is not and never has been a scheduling methodology. Unfortunately many organizations have placed emphasis on Lean as if it could solve scheduling problems by eliminating waste in capacity, thereby reducing the stress created by inadequate scheduling systems.

There is a direct relationship between lean effectiveness and reductions in material levels and decreases in capacity demand when action is taken to shorten setup time etc. In this way Lean technology is helpful for the both ends of the spectrum for point solutions. But lean management effectiveness rapidly deteriorates as conditions move toward less repetitive HMLV conditions. However, technicians continue to attempt to force Lean solutions to work for a HMLV environment requires high precision system scheduling solutions. This does not imply that lean solutions are not appropriate for managing some conditions in HMLV applications that benefit by reducing setup and other point solutions; it does mean that lean solutions are not as effective for capacity scheduling applications in HMLV environments. But, in reality, lean solutions are not a substitute for FCS in either a HMLV or a HVLM environment. In summary, most capacity management is more complex than material management and therefore production control for these cases should be primarily driven by capacity considerations, with material as the secondary constraint.


When considering the question “Why schedule?” it is necessary to focus on four universal business success factors:
1. Quality
2. On-time delivery
3. Price
4. Customer Satisfaction

Scheduling impacts all these areas. American industry today is striving to improve its efficiency and effectiveness and to shorten product cycle times. Several strong fads include Lean (from the Toyota Production System – TPS), Six Sigma, Total Quality Management (TQM) and Business Process Re-engineering (BPR). The Holy Grail is reliable production in the shortest time. The measure of success is almost always time-based.

The first time-based measure is predictability—the ability to deliver when quoted. Thus we get the TQM mantra of “Say what you do, and do what you say.” The second measure is cycle time reduction. This is the shortening of the time needed to produce an order. To accomplish these two time-based criteria requires capacity management, which requires effective scheduling. As Peter Drucker persuasively argues, the new measure of competitive success is, or should be, time. The true measure of productivity is output per unit of time given finite resources.

The time-based objective of scheduling should be to define when jobs will be completed and to deliver the jobs on that schedule; to deliver the product when promised. The cost of late delivery is high. At worst, it means lost orders or even lost customers, but it often also means excess inventory with high buffer inventories, poor customer relations, and excessive expediting. The bottom line costs are easily calculable. In contrast, the ideal schedule will:

Maximize resource utilization.
Decrease inventory costs.
Increase inventory turnover.
Improve customer service.
Improve communication and coordination.
Produce WC to-do lists that can be followed.


A company strategy should address conditions that promote company objectives; these objectives are modeled in the scheduling system to support adherence to the company strategy. Some important issues, which have often been left out of a company’s strategy, include:

1. Methods of quoting due dates.
2. Periodic evaluations of the strategic plan.
3. Sufficient attention to company wide communication (data accessible to all personnel when and where needed).

Failure to define and follow each of the points at all levels will tend to undermine the success of a scheduling project. The end result will be a tendency to deviate from good scheduling practices and lose its intended effectiveness of the FCS system.

Even the best companies will be reactive because of poor scheduling. Often their employees prefer being reactive because they are so used to operating in this mode. The “shoot from the hip” approach is very common in the American culture. The West was won with a six-shooter, and many shop floor managers continue to function as if every day is high noon in Dodge City. While there are some conditions that might benefit from this philosophy, modern, competitive, and complex manufacturing is not one of them.

This book will repeatedly point to the scheduling inadequacies of MRP systems and it may appear at times that we think MRP systems are of little use. This is not the case; to set the record straight, MRP systems and their mother ship Enterprise Resource Planning (ERP) systems contribute greatly to the material management function. But as an operations management execution function, MRP systems are totally inadequate and produce unfeasible results due to the assumption that capacity resources are infinite.


The authors’ experience in implementing FCS systems has resulted in some observations. One of the most discernible and useful observations is that practically all scheduling management prior to FCS has been based on a style that we refer to as job management. Wherever we implemented FCS, regardless of how the product was being manufactured, the occurrence of expediting was frequent. After several years of observing excessive expediting, it occurred to us that the problem was related to the inability to predict finish dates of customer orders. Finish dates were often later than due dates or would require expediting and/or over time to meet the established due date commitment.

Observing this high level of expediting led to the realization that several factors combine to create this perpetual condition. The process of piecing together the various influences on production and delivery commitments led to the conclusion that infinite capacity scheduling was the primary problem. However, infinite scheduling is not the sole contributor. The effort to satisfy clients in the absence of accurate data was also a large contributor to the chaos. And the acceptance of standard cycle times also contributed.

Most companies would apply some standard delivery time (predetermined product cycle time) for each product and quote the standard regardless of the load in the shop. To further exasperate the condition, because companies like to be responsive to client needs or demands and because of pressure from the sales department, the manufacturer would often quote whatever delivery the client insisted on, and then rely on expediting to resolve the problem.

Infinite scheduling systems assume that an operation can be started whenever it arrives at a work center. This is an incorrect assumption. In an attempt to correct this assumption, estimated queue time is added to the standard production lead time at each work center. This is time above what it would take to complete the product if infinite capacity did exist. This estimated time is based on some average and does not take into account the actual current demand. Ignoring variable demand is a second incorrect assumption. Other erroneous assumptions also occur, however, the following two major problems are both based on the assumptions:

A task can be started whenever it arrives at a work center.

The average queue time can be expected.

Often when a customer is on the phone ordering a product, they are requesting a delivery date that is earlier than the date calculated by the planning system. Since everyone is aware that the system is not accurate and that the estimated dates are buffered and therefore are much longer than the actual work content, the client gets promised the date requested. If this were the only exception to the estimated cycle time, there would be no problem because with infinite capacity systems these estimates are always conservative and much longer than normal delivery lead times. However, when the next customer calls with a delivery date request and also gets the delivery date desired, the problem is compounded. The result is that the exception becomes the standard and soon practically all jobs end up needing to be expedited if they are to meet the due date. This is an extremely costly method to run a manufacturing facility.


Most companies that implement software of any variety, be it accounting, forecasting, or material control, use only a portion of the functionality that the software system has to offer. They do not realize the total capability of the software. Scheduling software follows the same pattern. Many users continue to resort to intuition rather than using the system’s capability to find the best solution.

To utilize modern scheduling technology requires deviating from traditional practices. For example, some production managers are schooled that the bottleneck work centers become the major influence in limiting throughput. Regardless of what decisions are made relative to priorities, setup minimization rules used, and so forth, the controlling factor is that bottleneck work centers have more demand than capacity available. The underlying assumption is that the bottlenecks are static and unchanging, and in many applications, particularly in a job shop environment, these constraints often move from work center to work center (wandering bottlenecks).

Using an FCS system requires experience and has a learning curve prior to reaping the rewards. This requires a concerted effort. Intuitive methods should remain as a guidepost in learning a system. However, the major emphasis should focus on making use of the power of the scheduling software. Intuition resulting from years of experience should be used in con junction with the FCS system capability, such as what-if scenarios, to investigate options. However, the user should recognize the trap of “the schedule doesn’t look right.” Scheduling software that is truly analyzing the constraints to find a solution may take a path that the user would not recognize. This comes from years of instinctive “pattern recognition” of past manually-generated schedules that seemed to have worked out.


Why FCS; because reliance on infinite loading methods are unfeasible, inaccurate and costly. More importantly, reliance on these methods is unwarranted because computer technology is now more than capable to run finite capacity scheduling software. There are no longer any excuses for not seriously considering FCS to manage production on the shop floor. FCS is an integral capability to be included with the tools that are currently available to companies seeking to become world class, as indicated by the number of prominent ERP vendors that claim to have implemented FCS.

“Common sense is not common.”
— Will Rogers

Written By: Bill Kirchmier