Professional Documents
Culture Documents
220 - 1 Leading Indicators Execution Phase
220 - 1 Leading Indicators Execution Phase
220 - 1 Leading Indicators Execution Phase
LEADING INDICATORS
DURING PROJECT EXECUTION
Prepared by
Construction Industry Institute
Leading Indicators to Project Outcome Research Team
CII members may reproduce and distribute this work internally in any medium at no cost
to internal recipients. CII members are permitted to revise and adapt this work for their
internal use provided an informational copy is furnished to CII.
All CII members, current students, and faculty at a college or university are eligible to
purchase CII products at member prices. Faculty and students at a college or university
may reproduce and distribute this work without modification for educational use.
Chapter Page
Executive Summary v
1. Introduction 1
2. Methodology 3
3. Findings 12
4. Application/Implementation 16
References 23
Executive Summary
Over the years, the Construction Industry Institute (CII) and others
have conducted significant research to analyze and measure risks during
project development, resulting in several risk analysis and prediction
tools. Perhaps the best known of these is the CII Project Definition Rating
Index (PDRI), which is used by owners and contractors during front-end
planning. Once the PDRI exercise is completed, most tracking during
project execution is accomplished with traditional measurements. What
is needed is a method, once the front-end planning and risk analysis
have ended, of identifying indicators that most likely will lead to project
success.
v
1
Introduction
Objectives
The research project had the following major objectives:
1. Define a “leading indicator” and formalize the scope of this
project.
2. Develop a list of possible measures and means for evaluation.
3. Develop a prioritized list of measures and a weighting system
based on the statistical correlation of these indicators to project
health and phase.
4. Incorporate the measures and weighting system into a tool that
may be used to evaluate projects.
5. Develop a management process for implementing periodic
project reviews based on composite and individual scoring
systems.
A tool that does not rely on hard data and allows projects to identify
issues earlier is a valuable supplement to traditional tools. The research
team developed this new tool, called the Project Health Indicator or PHI,
which has broader application because management can utilize it at all
levels of the organization to assess project “health.”
1
The research team also wanted the tool to give some guidance regarding
project characteristics that lead to project risks to enable a starting point
for mitigation strategies. The method chosen was to relate the leading
indicators to identified project practices. Therefore, in addition to
identifying risks to project outcomes, the PHI tool shows the potential
project practices that could be utilized to improve the project risks.
2
2
Methodology
3
3. Quality/Operability: these are outcomes that are based upon a
facility being capable of operating per its intended function and
that the quality of the facility and construction craftsmanship
matches the intended asset life.
4. Safety: Safety as an outcome is a combination of construction
safety during the course of the project and overall safety
considerations of the new facility that will enable it to operate
safely over its life cycle. Construction safety involves accidents
to personnel in the battery limits of the construction zone and
is generally viewed in terms of recordable cases or Days Away
or Restricted Time (DART) cases. Facility safety, a more long-
term outcome, is based upon the facility having the equipment,
protections, and/or warning/safety devices, safe job procedures,
and energy control procedures required for the facility to operate
and be maintained in a safe manner.
5. Stakeholder satisfaction: Stakeholder satisfaction is the overall
pride, contentment, and/or happiness that the stakeholders have
with the outcome of the project. For contractors, it is somewhat a
measure of the potential for future, repeat business.
4
The original 180 potential leading indicators (LIs) were evaluated using
three criteria:
1. The problem the LI addressed
2. The project outcome the LI had an impact on
3. Whether the LI could be measured using a tangible or intangible
measurement means.
This initial screening effort reduced the number of LIs from 180 to
126. These were then ranked using a survey that assessed the potential
negative impact of each LI on the five project outcomes. This survey
was completed internally by 15 research team members. The results
were analyzed statistically based on a LI’s total negative impact on each
project outcome. Each LI was ranked from highest to lowest impact.
Based on the rankings and further evaluation, the number of LIs was
reduced to about 90.
A third survey was conducted to reduce the LIs to 43. This survey
included all CII membership. The results are discussed later.
This process of identifying and reducing the number of LIs was by far
the most significant effort in the research project and each measure was
carefully scrutinized, worded, and reworded in order to ensure that the
indicator represented a leading — not lagging — measure. The overall
goal was to ensure the tool had adequate measures to forecast results,
but to keep measures limited to avoid making the tool too cumbersome.
Traditional Measures
One dilemma was the problem of traditional measures versus this
new approach to identifying leading indicators. Obviously, owners and
contractors both spend significant effort and resources on traditional
5
methods of project control. Although not always effective, these methods
have led to improved project outcomes. Rather than simply ignore them,
the research team opted to include and scrub them along with the other
leading indicators. In the end, however, most traditional measures did
not meet the agreed-upon definition of a leading indicator and were
dropped from the refined list.
6
• TIC/Process Equipment
• Percent of Engineering Complete at Start of Construction
• Productivity and Forecasting for Engineering/Construction
• Source of Estimate (owner, contractor, third party)
• Number of Holds on Construction Drawings (quality of
authorized for construction documents)
• Cost Reports (committed cost/spending)
• Milestone Variances
• Cost of Hourly Rate vs. Plan (engineering and construction)
• Contingency Used (start of construction, reporting periods)
• Procurement Status (expediting)
• Submittal Status Report (vendor drawings)
• Overtime Percent
• Accounts Receivable
• Craft Turnover/Availability
• Earned Value = Percent Spend/Percent Budget
• Estimated Quantities vs. Actual Bulks
7
that the project has been sufficiently developed for execution. It should
be noted that if this assumption is incorrect, the PDRI questions should
be a primary source of potential execution risk.
8
Change Management: CII and others have accumulated large
amounts of research regarding the effects of late project scope
changes and high volumes of rework to poor project outcomes.
How the project team makes decisions on, controls, tracks, and
implements changes on a project can have a significant effect
on project outcomes.
Constructability: Constructability (construction input to design)
generally involves construction-related methodology and
planning. The ability to plan and execute the construction of a
facility efficiently is a major driver behind project success.
Contracting: Contracting is based on matching contract types
to project risks. It is not an endorsement of any one particular
contract type. There is no weighting of the PHI tool that values
turnkey versus lump sum versus design-build versus cost
reimbursable. It is purely a measure of whether the project team
is seeing potential issues between the contracts in place and the
scope that needs to be executed.
Quality Management: Quality management includes items such
as quality of engineering, construction quality and rework,
equipment inspections and testing, and facility start-up.
Safety Practices: This is a measure of whether or not the project
team is fully engaged in the practices that drive project safety
(see CII Target Zero practices).
Project Control: Project control involves the tools and techniques
used to track, evaluate, and improve schedule and cost
performance. In terms of this tool, it is not simply the use of
a project schedule and cost reporting. It is a measure of how
accurate the schedule is; how effective the schedule is in
tracking work and identifying gaps; whether the cost reporting
is utilized in future decision making; and whether the team is
effectively using the information as a planning tool. Too often,
schedules and cost reports become deliverables themselves
instead of tools to be used in planning the work.
9
Team Building: People implement projects. The core
competencies of the people that constitute the project team and
how the people that make up the project team play a key role
in the success of any project. Good project teams overcome
gaps in scope, risk events, design issues, and project changes
in a proactive way to minimize the negative effects on project
outcomes. Poor teams do not.
Some discussions were held regarding whether the tool should be the
same for owners and contractors or whether separate tools would be
more appropriate. Although the individual scoring between owners and
contractors varied somewhat, statistical checks were made to ensure that
the differences between owner and contractor scores would not bias
the results. Thus, the overall ranking of the leading indicators was not
substantially affected. Therefore, to allow the PHI tool to be used in a
team environment, the research team decided to develop only one tool.
10
The research team also decided to normalize the scoring for each
outcome and practice to 1000 points. This was done to remove the
scoring from leading indicators that were answered as “Not Applicable”
from the scoring system. A “Not Applicable” answer in the tool removes
the scoring for that LI from the calculation and the remaining scores
are then normalized around a maximum of 1000 points. In this way,
answering “Not Applicable” has a different effect from answering
“None.” An answer of “None” keeps the impact in the calculation. Thus,
the tool does not give credit for an LI that is not applicable to a particular
project and recalibrates itself to the leading indicators that can have an
impact on project risk.
11
3
Findings
12
Weighting
During the testing of the tool, an issue arose regarding how the weighting
magnified differences between LIs. As noted earlier, aggregated scores of
each LI for outcomes became the basis of calculating the impact of LIs on
outcomes. The base or normal weight for the LIs in the tool was derived
by dividing the aggregated score of a LI by the standard deviation of the
survey answers for that LI. This determined the maximum points assigned
to each LI. Thereafter, the team assumed a linear relationship for the
problem levels in the tool and, therefore, the model simply assigned a
multiplier of serious (maximum points) = 100%, major = 75%, moderate
= 50%, minor = 25%, and none = 0% to the weighted scores to quantify
the risk impact of the LI.
Two effects were observed. First, the risk scoring for the outcomes
tended to be similar. Since only high impact LIs were included in the
tool, the range in the LI weights across all outcomes was not that great.
This caused some concern since the results did not always highlight the
outcomes that may be at a higher risk of poorer performance. Second, it was
noted that the normal weighting (dividing by a standard deviation) often
did not create substantial differences between outcomes. Thus, alternate
weighting methods were explored that would produce differences in the
risk assessments. That is, dividing by the standard deviation might not
be significant enough to differentiate between outcome scores. Based
on these two issues, the research team concluded that it needed to test
different scoring weights.
In the evaluation of the project data, a version of the PHI tool was
created that allowed users to choose one of five different weighting
options (created by modifying the standard deviation). The users were
13
asked to select the weighting option for the tool that best reflected the
health of the project being evaluated. In the tool, the higher weighted
options tend to emphasize the LIs that have consistent answers (small
standard deviations) and de-emphasize the LIs that have inconsistent
answers (large standard deviations). Statistical correlations were evaluated
between each of different weight options and the project outcomes.
Overall, dividing by the fifth power of the standard deviation had a
higher correlation and was therefore selected as the scoring method of
choice for the tool. The scatter plots of the project scoring using the
selected weighting method versus the cost and schedule outcomes are
shown in Figure 1.
14
Cost Correlation
Series 1 Linear Regression
1000
900
Project Cost Health Score
800
700
600
500
R2 = 0.6798
400
300
80 90 100 110 120 130
Performance (% - Actual / Budget)
Schedule Correlation
Series 1 Linear Regression
1000
900
Project Schedule Health Score
800
700
600
500
400
R2 = 0.5976
300
80 100 120 140 160
Performance (% - Actual / Budget)
15
4
Application/Implementation
The user simply reads each LI statement and selects the potential
problem impact that a particular indicator may have on the project.
Generally speaking, it requires less than 30 minutes for someone familiar
with the tool to read and answer the questions. The scoring criteria is
embedded in the tool to assist the user with quick guidance should
questions arise concerning, for example, the definition of “serious”
versus “major” as shown in Figure 2 for LI 3.
16
During development, several options emerged as ways to fill out the
tool. Many organizations had the project manager fill it out. Some had an
engineering manager with oversight responsibility fill it out while some
did it as a group exercise.
17
Figure 3. Example Project Health Outcome Gauges
with Normal Weight
18
Again, these scores are normalized to 1000 possible points, with a
lower score being interpreted as higher risk to that outcome. It is important
to note that the scores are a measure of potential risk of not achieving
desired project outcomes. This view is different than that provided by
traditional project controls. The PHI tool is predicting risks based on the
evaluation of non-quantitative information. It is an evaluation regarding
the likelihood of future events effecting outcomes. Also, in addition to
the five project outcomes, a composite overall health score is included.
19
Similar to the project outcome risk scores, the practice correlation
scores are normalized around 1000 possible points, with lower scores
being considered not desirable. Unlike the project outcome scores that
are a measure of risk, the practice correlation scores are an indicator
of deficiencies or gaps in the project practices that are leading to the
outcome risks.
The above only adjusts coloring of the gauges on the output screens. It
has no effect on the scores. It is provided purely as a means to adjust what
levels of risk are considered acceptable to an individual organization.
Some examples for reasons to adjust the gauges could include sensitivity
to safety issues or schedule compression. Moving the colors of the gauge
has no effect on the risk assessment of the project by the tool and the
correlations of the data to project outcomes.
20
Figure 6. Input Form for Calibrating the Outcome Gauges
21
5
22
References
Choi, J., Anderson, S., and Kim, T., “Forecasting Potential Project Risks
through Leading Indicators to Project Outcome,” Research Report
220-11, a report to the Construction Industry Institute, The University
of Texas at Austin, in press.
Hawley, R., and Hawley, D., Excel Hacks: 100 Industrial-Strength Tips
and Tools, O’Reilly Media, Cambridge, MA., 2004.
23
Notes
25
Leading Indicators to Project Outcome Research Team
* Stuart D. Anderson, Texas A&M University
Robert H. Briest, ALSTOM Power
David Campbell, Air Products
* Jiwon Choi, Texas A&M University
William Chumchal, Mustang Engineering
* Paul Ennis, U.S. Steel, Co-Chair
Ian D. Etzkin, CDI Engineering
Gilles L. Gelinas, DuPont
* Kevin J. Gierc, Dick Corporation, Chair
Stephen E. Goodman, J. Ray McDermott
Robert MacLaren, Ontario Power
Michael G. Pappas, Eastman Chemical Co.
Steven D. Rench, ConocoPhillips
Mike Sheridan, International Paper
Brian E. Sichter, CB&I
Past Members
Michael Bronkowski, Amgen
S. J. Kim, Texas A&M University
Minesh Kinkhabwala, Aker Kværner
Don G. Miller, Gilbane
* Principal authors