MANAGING THE CRITICAL PATH/CHAIN IN INSTRUCTIONAL DESIGN

 

 

 

 

By Arnie Witchel

 

Copyright 2003: Witchel & Associates

 

 

 

 

 To return to the Publications, click here:

 

 

 


Abstract

 Instructional design is the process of determining objectives, strategies, techniques and materials necessary to meet instructional goals. As such, the theory and production skills used in instructional design should be matched with project management skills in order to be effective and efficient in the production of high quality instructional design products. At the heart of project management is the concept of the critical path. While some literature has been devoted to the importance of project management skills for instructional designers and the use of project teams in instructional design, little has been done to further the application of the latest insights of the critical path method, including the critical chain process, to the process of instructional design. This paper reviews the literature in the field and makes suggestions for practical application to instructional designers.


The Problem

     Instructional design is the process of determining objectives, strategies, techniques and materials necessary to meet instructional goals (Gentry, 1994). As such, the theory and production skills used in instructional design should be matched with project management skills in order to be effective and efficient in the production of high quality instructional design products (McDaniel & Liu, 1996). At the heart of project management is the concept of the critical path. While some literature has been devoted to the importance of project management skills for instructional designers (McDaniel & Liu, 1996) and the use of project teams in instructional design (Koller & Frankenfield, 2000), little has been done to further the application of the latest insights of the critical path method to the process of instructional design.

Review of the Literature

     The Critical Path Method (CPM) and Program Review Evaluation Technique (PERT), were developed by the Navy in the 1950s to aid in the development of the Polaris Missile programs (Freeman, 2000), a program that involved 250 prime contractors and more than 9,000 subcontractors (Defense Systems Management College, 1990). The CPM involves “the sequence of interdependent activities that drives a project’s schedule and tracks those activities that logically must precede and follow a given task” (Loulakis & Santiago, 1998).

     While often lumped together as PERT/CPM, the two concepts have computational similarities, but also significant differences. PERT uses a probabilistic view, allowing the analyst to see the effects of randomness and uncertainty on the project completion; any variances in PERT work completion are strictly random, beyond the control of the project manager (MacLeod & Petersen, 1996). CPM, on the other hand, uses a more deterministic view of the project: The project manager makes initial estimates of the task completion times and costs, then later makes further resource allocations if the initial estimates lead to unsatisfactory results (MacLeod & Peterson, 1996).  The constraints associated with the critical path are implied by predecessor-successor links as well as relationships such as start-to-start and finish-to-finish, calculating the earliest date a successor begin may begin following the earliest date a predecessor activity is concluded on the forward pass, and the backward pass calculates the latest finish of a predecessor task is determined by the earliest late start date of the successor activities, accounting for lag time (Thomasen & Butterfield, 1993).

     However, while different in approach, historically both share a common calculation method of the amount of time it will take to complete a project. PERT advocates an approach that statistical uncertainties are applied as optimistic, most likely and pessimistic time for the task durations. These act in the place of the mean of the project (most likely time) and the earliest and latest start and finish times.  However, to attain statistical parameters, Van Slyke in 1963 advocated using the Monte Carlo Simulations (Thomasen & Butterfield, 1993). In this method, a series of CPM schedules using random distribution to slightly change the change the activity durations, followed by normalizing the data, lead to a stochastic network, and the popular equation used to estimate the length of tasks today: o+4m+p/6; or the optimistic time, plus four times the median or most likely time, plus the pessimistic time, divided by six. While this is in wide use in project management at this time, it does have limitations in estimating parameters because the cumulative distribution function is assumed to be normally distributed based on the Central Limit Theorem. Keefer and Verdini (1993) suggest that this can lead to large errors in the cumulative distribution function and have found two formulas that offer more accuracy, with a stronger recommendation for the Extended Pearson-Tukey option. In further work, Golenko-Ginzberg (cited in McLeod & Petersen, 1996) has concluded that trying to determine the value of m (the mean time) is practically worthless and has offered still other formulas to improve forecast accuracy. Still, these formulas are not widely accepted in project management circles. Leach (1999) points out that while refinement of time estimates is important, it doesn’t add to the ability to manage the actual project.

     In terms of its application to instructional design, these methods of calculation can be especially problematic. Unlike projects such as construction, instructional design is a service and often relies on knowledge gain and translation to other forms in constructing the design package. Time estimates for designing on topics with which the designer is intimately familiar and on topics that are only somewhat familiar can vary widely, and depend on a great number of factors including types and availability of materials available. In construction, the CPM is sometimes used to prove the time and cost impact of a delay, and it has been used by contractors as a way of identifying and proving delays for claims purposes (Loulakis &

   Due to the differences in approach (random vs. deterministic), the remainder of this paper will focus on the CPM and its use in managing projects, particularly instructional design projects, as well as further refinements in the project management body of knowledge regarding conceptualization and application of the nature of restraints and management. The value of the CPM is that it allows the project manager to identify the migration of critical elements (Beckman, 1999). If approached using dynamic predecessor time considerations, it can offer opportunities to shorten the critical path by scheduling tasks in concurrent, parallel paths (The Hampton Group, 2001). However, a major complaint about CPM is that it is based on management by exception, which involves identifying and isolating critical or conflicting information (Rahbar, Yates & Spencer, 1991). The argument against CPM as a tool is that it monitors only time for evaluation of project evaluation and fails to monitor achievement of objectives because verification of objectives (which differ from goals, or specific tasks) are difficult to verify (Rahbar, Yates & Spencer, 1991).  However, somehow this argument regarding objectives seems to vary from the definition of a project as “A temporary endeavor undertaken to create a unique product or service” (Project Management Institute Standards Committee, 1996, p. 167). It also fails to note the purpose of the CPM approach, which is to achieve the overall project goals, which are linked to the objectives of the organization. It is important that the project manager optimize the critical path with regularly updated estimates of work and task duration to manage in front of the team, watching the whole critical path, and not just the in-process tasks (The Hampton Group, 2001).

     Perhaps the actual complaint is that the CPM focuses so much on the critical path that it may not focus on the environment in which the project exists. Another approach that combines CPM with the environment is Critical Chain Project Management (CCPM). CCPM reengineers project management to remove common problems that lead to poor project performance (Leach, 1999). CCPM departs from traditional management of the critical path by specifying the critical chain, rather than the critical path, as the project constraint, including resource dependencies, and does not change during the project execution. Constraints in this context may be timing problems, resources, market demand, raw materials or even management policy (Thomas, 1997).  In addition, it uses best estimate, or 50% probable activity times, and adds buffers at the end of activity chains to allow for uncertainty in estimating and execution. It uses the buffers as a direct measuring tool, defining the constraints for multiple projects as the constraining company resource (thus linking to the company objectives) and seeks to change team behavior encouraging early performance of activities (Leach, 1999). Following on his 1984 work on the Theory of Constraints (TOC), in which he noted that all systems must have constraints or they would grow without bounds (Leach, 1999), Dr. Eliyahu Goldratt developed the critical chain method (CCM) in 1997 (Michalski, 2000). Unlike CPM, which uses a pessimistic, optimistic and most probable time estimate in measuring task duration, CCM squeezes most of the safety out and concentrates solely on the tasks that are critical to the project’s success and upon which other tasks are dependent. The resources that are required in the chain are identified, and multitasking is minimized in an effort to more quickly complete the task in the chain. Kania (cited in Michalski, 2000) likens this to a baton race, where the moment the baton is relayed, all resources are focused on completing that leg of the race of the runner who holds the baton. However, the critical chain method does employ buffers to serve as a safety net. These buffers are added and removed as needed, and serve as a marker as to which activities use the most buffer consumption and which may be removed and added to another task. The buffers are used to protect what Deming (cited in Leach, 1999) called Common Cause Variance, or variance that arises from the system (as opposed to Special Cause Variance).

     CCPM was developed from systems thinking and the Theory of Constraints (Leach, 1999). TOC’s primary premise is that there are built in constraints in any project that need to be identified and managed, or the business will suffer needless setbacks on the project (Michalski, 2000). Constraints not only limit the operation’s capacity and the efficiency of the organization, but a constraint can be systematically upgraded until another takes its place because production improvement is a continuous process (Thomas, 1997). TOC’s ultimate objective is to establish a rhythm set by the constraint, known as the drumbeat. It is the buffer time that separates the constraint from the rest of production and protects the critical task from the rest of the operation that can cause possible disruptions (Thomas, 1997). Rather than synchronizing the critical path to a predetermined length, TOC seeks to synchronize all other operations to the rate of production of the constraints, forming the concept of rope, and forces all parts of the development system to work at the pace set by the drum; therefore, the terms drum, buffer and rope refer to TOC (Thomas, 1997).

     While TOC has wide application in the Manufacturing Resources Planning arena, it has also been successfully used in small businesses and in the military, to improve health plans. While constraints can be classified by cause, Womack and Flowers (1999) assert that many constraints are the result of organizational rules, training, or measures that act as policy restraints. Fewer constraints are resource constraints and even fewer are market constraints. Bushong and Talbott (1999) demonstrate that constraints may exist because of lack of scrutiny about throughput of product and yield analysis.

     Goldratt (1997) extends the production application of TOC to projects. He identifies the constraints of projects as the sequence of dependent events that prevents the project from completing in a shorter interval; resource dependencies as well as activity dependencies determine the critical chain of a project. Rather than focusing on the critical path per se, the TOC focuses on the critical chain as well as the project goal and the constraint that blocks the achievement of the project goal. There are five steps in successfully achieving the project goal: Identify the constraint, exploit the constraint, subordinate everything to the constraint, elevate the constraint, and repeat the process if a new constraint is identified (Leach, 1999). This is used to keep focus on the goal and to eliminate a trap in CPM: Focusing on date driven behavior, rather than the critical chain and completion of the project.

     It is interesting to note that this approach of focusing on the critical chain takes away many of the problems with traditional time driven project management. The problems of slack and crashing the project (Freeman, 2000) are removed, because buffers are built in and adjusted as needed and performers that run over estimated activity durations do not face criticism as long as they start the activity as soon as they have input, work 100% on the activity, and pass on the activity as soon as completed (Leach, 1999).

     This reinforces the importance of queuing the project for immediate start, as well as passing it on in a timely manner. This would be especially true in instructional design organizations, where topic familiarity might influence the duration of task activities, and multi-project designs would influence the overall project completion. Queuing often has a significant impact on a project’s delay in completion as well as an impact on costs; however, applying queuing theory to the critical chain can reduce those costs and allow the focus necessary to start the activity as soon as it is received and pass the activity on as soon as it is completed (Levy & Globerson, 1997). Like Just-in-Time, queuing can set the rate of production by constraint, rather than demand, by establishing the rhythm of production (Thomas, 1997).

Application of the Literature in Instructional Design

     Instructional design can benefit from the literature reviewed. Most instructional design work is done on a project basis. However, many instructional designers are unfamiliar with the concepts of critical path and critical chain because the management processes for producing a project are not always readily apparent or available to instructional design students (McDaniel & Liu, 1996). In addition, many instructional designers are not professionally trained in project management. One of the problems of application of the CPM to instructional design is the uncertainty of service work, especially knowledge work. Even if time parameters were defined, it would remain an estimate at best, reinforcing Golenko-Ginzberg’s assertion that the value of the mean time is practically worthless (cited in McLeod & Petersen, 1996).  However, the critical chain in instructional design projects, as well as the resources needed, are readily identifiable, as are constraints to project completion. Any instructional designer or manager of instructional design projects can identify resource dependencies, as well as activity dependencies that determine the critical chain of an instructional design project. The application and knowledge of constraints can allow the constraints to be identified and exploited, as well as subordinating all else to the system constraint and elevating the constraint (Leach, 1999). Constraints in instructional design can include the timing of when the design needs to be completed, resources available, market demand for the designed product, materials or even client policy regarding the design. Brainstorming to identify constraints and solutions to constraints can improve the design team’s success in handling constraints (Womack & Flowers, 1999). Scrutinizing the throughput and yield of design projects can result in better return on investment for the design team and help to rearrange priorities (Bushong & Talbott, 1999). Identifying and communicating client policies or measures that constrain quality and timely delivery can also lead to more successful partnerships (Womack & Flowers, 1999).

     The use of realistic time parameters and buffers to allow for common cause variance is also applicable to an instructional designer’s situation and tools. The buffers can be moved and arranged to concentrate on the task in the design project that needs more resources for successful project completion (Michalski, 2000). They give the designer and design team more flexibility by not focusing on date driven frameworks.

     Use of these methods cannot only improve instructional design projects, but they can also be used as a communication tool with clients. If a customer requests a change in an activity that will impact the critical chain of dependencies and resources, the instructional designer can demonstrate how this will impact the and delay the project completion and increase costs (Beckman, 1999). Instructional designers would be wise to teach their clients the concepts behind the critical chain and critical chain management. In this way, the client can use the critical chain as a requirements and date planning tool, and to develop their own timeline as well as identify possible constraints that they can influence, both positively and negatively (Beckman, 1999). As Beckman (1999) advocates for the CPM, the critical chain can even be used as a negotiation point by instructional designers and their clients. By better understanding the concepts behind the CCPM, instructional design teams working together can better scrutinize the critical chain, improve queuing procedures, reduce multitasking and improve the use of buffers by better defining common cause variation.

     Compared to the CPM, CCPM planning and project management is simple, without three time estimates, Monte Carlo methods, and complicated statistical analysis (Leach, 1999).  Like CPM, however, it can improve problems with time delays, cost increases, and customer communication, while delivering a better instructional design product. Instructional designers need to identify constraints that can cause project delay or affect project quality, learn to improve them, and increase their ability to exploit restraints, subordinate everything to the constraint, elevate the constraint and return to identifying constraints. This will improve their critical chain management, while at the same time improving the quality of the project design.

References

Beckman, I.P. (1999). Delivering high-quality products with the CPM. Engineer, 29 (3), 48-50.

 

Bushong, JG., & Talbott, J.C. (1999). An application of the theory of constraints. The CPA Journal, 69(4), 53(3).

 

Defense Systems Management College. (1990). Scheduling guide for program managers. Defense Systems Management College: Fort Belvoir, VA.

 

Freeman, L.N. (2000). Two planning tools can help you manage projects: Gantt charts, critical path method graphs are powerful scheduling aids. Ophthalmology Times, 25(18), 18.

 

Goldratt, E.M. (1984). The goal. New York: North River Press.

 

Goldratt, E.M. (1997). Critical chain. New York: North River Press.

 

Keefer, D.L., & Verdini, W.A. Better estimation of PERT activity time parameters. Management Science, 39(9), 1086-1091.

 

Koller, C.A., & Frankenfield, J.J. (2000). Twelve tips for developing educational based multimedia in a community-based teaching hospital. Medical Teacher, 22(1), 7(4).

 

Leach, L.P. (1999). Critical chain project management improves project performance. Project Management Journal, 30(2), 39-51.

 

Loulakis, M.C., & Santiago, S.J. (1998). CPM schedule analysis insufficient to prove delay claim. Civil Engineering, 68(1), 43.

 

MacLeod, K.R., & Petersen, P.F. (1996). Estimating the tradeoff between resource allocation and the probability of on-time completion in project management. Project Management Journal, 27(1), 26-33.

 

McDaniel, K., & Liu, M. (1996). A study of project management techniques for developing interactive multimedia programs: A practitioner’s perspective. Journal of Research on Computing in Education, 29(1), 29(20).

 

Michalski, L. (2000). Applying the theory of constraints. Pharmaceutical Technology, 24(9), 126-132.

 

Nino, L., & Globerson, S. (1997). Improving multiproject management by using a queuing theory approach. Project Management Journal, 28(4), 40-46.

 

Project Management Institute Standards Committee. (1996). A guide to the project management body of knowledge. Upper Darby, PA: Project Management Institute.

 

Rahbar, FF., Yates, J.K., & Spencer, G.R. (1991). Project management knowledge engineering system. Cost Engineering, 33(7), 15-24.

 

The Hampton Group. (2001). Critical path—the road less traveled: Optimizing project plans. Available: www.4PM.com

Thomas, M., Jr. (1997). Emerging technology: Production scheduling matures. IIE Solutions, 29(1), 24(5).

Thomasen, O.B., & Butterfield, L. (1993). Combining risk management and resource optimization in project management software. Cost Engineering, 35(8), 19-24.

 

Womack, D.E., & Flowers, S. (1999). Improving system performance: A case study in the application of the theory of constraints. Journal of Healthcare Management, 44(5), 397(11).

 

To return to the White Papers, click here: