Catch up on the Latest News with Barchan’s Blog

Point of View: Choosing a Project Delivery System (PDS) in a Multi-Project Environment

Posted by on in Uncategorized

Point of View:

Editor's Note:  Public entities approach choice of project delivery system from different points of view.  One approach is to eliminate inappropriate methods for a given project from the list of Six Key Delivery Methods in the MIT Framework (see Barchan Foundation home page), followed by selection from those that remain of a workable project delivery system, with definition of scope of work and contract requirements.  Another approach involves systematic analysis of project delivery systems with the goal of choosing the BEST PDS for a given project. 

Y.Y. Chih and Prof. C. W. Ibbs [head of the construction management program at the University of California at Berkeley] describe one such approach for selection of a PDS.

We look forward to your comments and thoughts.

Choosing a Project Delivery System (PDS) in a Multi-Project Environment

By Y.Y. Chih (Ph.D. candidate, Dept. of Civil Engineering, University of California, Berkeley)
and C. William Ibbs (Professor of Construction Management, University of California, Berkeley.)

1. A project delivery system (PDS) selection problem

As Barchan Foundation users know, there are numerous variants of the Six Key Delivery Methods (PDS) available for infrastructure development. A question regularly facing public agencies is: which PDS should be used to develop a specific infrastructure project? The answer to this question is critical because the right PDS can significantly increase project efficiency and success rate. Choosing a proper PDS is however a difficult task for a number of reasons.

First, a PDS decision is often made early in the project planning phase when only limited and imprecise project information is available. That information scarcity is involves goals, timing, budget, etc. of the project, roles of the participants, the surrounding economic environment, and so forth. Secondly, selecting the right PDS involves a choosing between the infinite variety of PDSs.

And thirdly, the existence of dependencies among projects in a multi-project environment further complicates the decision-making process.

To address these difficulties requires a systematic decision-support framework, which is the subject of research at Berkeley.  This article describes a decision-support framework for approaching the selection of a Project Delivery System.

2. A PDS selection decision-support framework

The overall decision-support framework is composed of five distinct stages and gates (as shown in Figure 1). Stages are where tasks are performed, and each of them is further broken down into a set of steps. Gates are checkpoints where outputs of each stage are discussed, validated and approved by decision makers. 


Stage 1: define project objectives and constraints

The first stage towards a PDS decision is to define project objectives/constraints in terms of four categories: time, cost, quality and others. A project’s objectives/constraints on time, cost and quality can be a range of values. For example, the total project duration is preferably 15 months and cannot exceed 30 months; and the total project cost is preferably $5 million and cannot exceed $10 million. Some examples of other project objectives are: to seek innovation opportunities throughout the lifecycle, to minimize impact on the environment, and/or to minimize inconvenience to the community and public.

Stage 2: search for PDSs

The second stage is to develop a list of PDSs through a market search, a historical search, or both. These PDSs are then examined for their compliance with statute or regulations. If the use of a PDS is prohibited by statute or regulations, it is eliminated from the list. If the use of a PDS is only allowed under certain conditions (e.g., requiring extra approvals), these conditions should be considered in the PDS selection process.

Stage 3: define evaluation criteria

In this third stage, a set of selection criteria that reflect the pre-defined objectives are defined. These criteria can be grouped into: performance (PERF) and PDS criteria. PERF criteria include the three prime project performance measures: time, cost and quality. With rare exception PERF criteria are the project owners’ primary concerns because they need to meet certain mandatory constraints. For example, a project has to be completed within a certain amount of budget and by a specified deadline. The project also needs to meet a list of specifications to fulfill its planned purposes. How a project performs on PERF criteria is affected by the characteristics of a PDS (e.g., how project phases are integrated) and several other factors (e.g., project dependencies, time-cost trade-offs and external factors).

To integrate these factors into a PDS decision-making process, PERF criteria (except quality) are measured on a quantitative basis. For instance, time is measured in days, months or years; and cost is measured in dollars. Criteria other than time, cost and quality are grouped and defined as PDS criteria. Some common PDS criteria are time certainty, cost certainty, flexibility, complexity, risk avoidance, price competition, and point of responsibility. The project performance on these criteria is largely determined by the characteristics of a PDS. Because PDS criteria are mostly qualitative in nature, they are measured on a relative scale. The relationships between these two groups of criteria and the PDS decision, the factors affecting a project’s performances on these criteria and how these criteria are measured are summarized in Figure 2.b2ap3_thumbnail_evaluation-criteria.jpg

Stage 4: evaluate PDSs on the two groups of criteria: PERF and PDS

Dealing with PERF criteria. The initial step to evaluate PDSs on PERF criteria is to estimate project time (T), cost (C) and quality (Q), given a PDS, based on information from past similar projects. This information serves two purposes: first, it is used to derive time-cost trade-off rules; and second, it is used as a basis for further consideration of project dependencies. The second step is to identify dependencies between the new project and other ongoing projects. Next, the effects caused by these project dependencies on the new project’s estimated schedule and cost must be identified, and necessary adjustments made by applying the predefined time-cost trade-off rules. Decision makers can also adjust the new project’s estimated duration and cost for other desirable time-cost trade-offs.

The overall process is shown in Figure 3. This process is repeated for each of the PDSs under consideration. The final outputs are the new project’s estimated durations, costs and quality levels under different PDSs. These results are presented in a form of a performance profile, as shown in Figure 4.


Dealing with PDS criteria. A weighed score approach is used to evaluate PDSs on PDS criteria. In this approach the criteria weights are first determined. Then the eligible PDSs are scored based on these criteria. Next, the scores for each PDS are multiplied by the criteria’s corresponding weights to achieve weighted scores. These weighted scores are then being summed to arrive a total weighted score. 

This result is also presented in a form of performance profile, as shown in Figure 5. The positions of PDSs on the line indicate their relative overall performances on PDS criteria.

b2ap3_thumbnail_performance-profile.jpgStage 5: choose a PDS.

PDSs’ performance profiles on PERF and PDS criteria are combined to produce an overall performance profile. One such example is shown in Figure 6. In a performance profile, a PDS that dominates all others on all criteria (e.g., PDS2 in Figure 6) is considered as the most appropriate PDS. If no single dominant PDS exists, decision makers can scale the performance profile to indicate the criteria weights. A weighted performance profile can help decision makers distinguish the differences between PDSs and identify the stronger PDS. If necessary, the overall performance profile can be future broken down to visualize PDSs’ performances on individual criteria. For example, the time criterion can be broken down to show a project’s planning, design, bid, and construction durations under different PDSs, as presented in Figure 7.b2ap3_thumbnail_overall-performance-profile.jpg

3. Benefits of the framework

This decision-support framework is beneficial to PDS decision makers in several ways. First, it guides decision makers, step-by-step, in evaluating alternative PDSs considering multiple criteria. Furthermore, it integrates the considerations of project dependencies and possible time-cost trade-offs into the PDS decision-making process.

This, all together, allows decision makers to choose a PDS on a more informed and systematic basis. As such, the possibility of an inappropriate PDS being implemented can be reduced. In addition, the presentation of PDS performance profiles also helps decision makers identify the criteria on which a PDS may performs poorly once it is selected. This provides decision makers the opportunities to develop appropriate proactive strategies.

For correspondence with the authors regarding this article:  contact Y.Y. Chih at


To learn more about The Barchan Foundation, Inc. or to post interesting public infrastructure projects to the Barchan Foundation data base, please go to .

Rate this blog entry:
blog comments powered by Disqus
Loading ...

Blog Archive


You are about to delete: .

Confirm Deletion Cancel