In a highly variable multi-project environment it is very difficult to correctly determine the workload that should be placed upon any particular resource. There is no question that putting too little work into the system will tend to starve key resources. And while there is a natural desire to keep resources busy, in particular the critical resources, overloading them usually results in unfavorable outcomes. For example, overloading resources often result in bad multi-tasking. Furthermore, when resources get overloaded, it is very difficult to determine how these resources should prioritize their activities among the various tasks needing their service. As a result, organizations usually experience significant degradation in performance. Project management requires projects to be completed on time and meet customer quality requirements; and these objectives become difficult to achieve when key resources are overloaded.
In an ideal setting, work schedules can be developed in advance, so that resources have just the right amount of work allocated to them at various points in time. Scheduling resources in this manner might work well in a variety of production setting where much of the work on scheduling resources assumes a fairly deterministic scenario in which the demand is relatively predictable and the resources work on fairly repetitive tasks. However, in the project world, demand is highly uncertain, workflow is quite unpredictable, and task durations have significant variability. Even the best-planned schedules become difficult to execute in this environment. And when many different resources are used multiple times in a single project and frequently shared between projects, any unexpected delay in a single task can cause significant ripple effects delaying one or more projects. Even a small delay in a task far away from a key resource can cause chaos in the complicated and interrelated schedules that exist in a project environment, and attempts to tightly schedule projects are soon abandoned.
This presentation outlines three specific steps to dramatically improve the performance of these organizations. In order to enhance performance of the system, we present a simple approach to
The proposed resource loading methodology provides the project manager the ability to decide whether or not to release new work into the system. The methodology does not look at exact work schedules, nor does it look at unique timing situations that could overload even the best-planned schedule. Rather, it looks at the backlog of work for the resources, noting that this backlog of work for a resource will generally not all be present in the immediate queue of work at that resource.
- determine how resources should be loaded,
- establish task priorities for each resource, and
- identify the appropriate level of reserve resources
In addition to varying the resource loading, we also evaluated the impact of Multi-Tasking and Prioritizing Tasks. We look at both a multitasking environment as well as a non-multitasking environment. In both situations we studied the impact of task prioritization on project value.
The third part of our study investigated the application of additional expert resources to the projects. This last test is more interested in reducing the impact of high-variability tasks on the performance of the system. Tasks in typical projects have a high degree of uncertainty in the estimated task duration. It is very hard to predict when or where this tail will happen. It could happen at any resource, not just at the most heavily used resource. We were therefore interested in answering the following question, "How can we best minimize the effects of the occasional long tail (task durations well above the median) with the least effort?” We found an effective solution in the form of Expert Resources.
The strategies proposed in this presentation were evaluated using a discrete-event simulation model. The simulations demonstrate that a 200 to 600 percent improvement in project value is possible, where project value is defined as the number of projects completed over a given period of time divided by the 90 percent probable flow time for projects completed during this period.
Reducing the resource workload, setting task priorities and providing expert assistance when needed can dramatically change the culture of large multi-project environments for the better.
- How to improve performance in a project management environment?
- What is the appropriate resource workload in a project management environment?
- What is the impact of adding reserve resources?
Robin Clark is a founder of the QMT Group, which offers ExtendSim simulation courses as well as coaching, modeling and consulting services.
Robin is the primary simulation instructor for Imagine That, the makers of the ExtendSim simulation software. He has developed several simulation courses, including simulation database techniques, simulation development environments, rate-based modeling and simulation modeling for health care.
He has also built numerous simulation models for clients in the areas of health care, pharmaceutical manufacturing, automotive manufacturing, consumer goods manufacturing and supply chains.
Robin has a B.S. in Physics and a M.S. in Management Science.