Performance Contracting for Construction

Performance Goals

Process

The basis of any performance contract is the set of performance goals that defines what the contractor is to achieve under the contract. Development of these goals is time-consuming, and needs to be a group activity within the agency. A goal development process is described in the figure and subsections below.

The project team followed a similar process (with the exception of not performing the final two steps) to develop a sample set of performance measures and goals. This sample set of goals is presented in the Sample Materials section following the description of the process.

The project team followed a similar process (with the exception of not performing the final two steps) to develop a sample set of performance measures and goals. This sample set of goals is presented in the Sample Materials section following the description of the process.

Initial Brainstorming Sessions - What is the Owner Agency Trying to Achieve with This Project?

In defining performance goals, the first internal stakeholder brainstorming session should be focused on defining what the Owner Agency is trying to achieve with the specific project.

For HfL efforts, the general goals have already been defined at a high level. These include improving safety, reducing congestion due to construction, improving quality, and improving user satisfaction. However, the challenge here is to define the specific goals that the Owner Agency will use to judge success on its project.

In performing this step, the Owner Agency should cast a broad net in the brainstorming session and keep asking, "What have we missed?" The Owner Agency should document all of the ideas that are put on the table, whether or not they seem like good ideas at the time. As the Owner Agency moves through the process, it can always discard ideas that do not fit.

The brainstorming ideas at this stage will form the basis of the performance goals. Remember to focus on WHAT to achieve, not HOW to achieve it. The HOW will be up to the contractor to determine.

Goal Format and Writing

Once the Owner Agency has its goal ideas from the initial brainstorming session, the Owner Agency needs to consider a number of format issues for turning the ideas into goals. The first is whether the goal set should be detailed or higher-level general goals. The Owner Agency would go the detailed route if it is looking to drive the contractor's work plan at the detailed level. The Owner Agency will have a good idea from the brainstorming session which way to go. If the Owner Agency had many detailed ideas, it would likely go the detailed route. If the Owner Agency had more general ideas, it would go the higher-level general route.

The next issue is whether the Owner Agency should make the goals subjective or objective. Objective goals are generally preferable, especially if there is a national standard process available for measuring performance. Some goals, however, will need to be subjective. That is fine as long as the rating/scoring process is well defined. This will protect both the contractor and the Owner Agency.

What Makes a Good Goal?

Once the Owner Agency has its draft goals down on paper, the question to ask for each goal is, "Am I SMART?" The acrostic is as follows:

• I - Is the goal something that the contractor can influence?
• S - Is the goal specific?
• M - Is the goal measurable?
• A - Is the goal achievable?
• R - Is the goal results-oriented?
• T - Does the goal have a time element?

If the answer is yes to each of these questions for a draft goal, then the goal is likely a good one. If the answer is no, consider revising the goal into a different form or deleting the goal. Two other questions to ask include:

• Has the goal ever been measured before?
• Is there existing infrastructure in place to measure performance?

These last two questions are not go/no go questions, but the answers can impact the amount of effort needed to evaluate performance.

The Owner Agency will also need to make sure that the goals do not conflict with their standard specifications. If they do, the Owner Agency can modify the standard specification language in the solicitation package.

Organizing/Categorizing Goals

Once the Owner Agency has its draft set of goals, the Owner Agency needs to consider how to organize/categorize the goals. A good way to categorize the goals is by benefit category such as safety, construction congestion, quality, user satisfaction, etc. Another way is to categorize by infrastructure type.

The Owner Agency should place the goals into the various categories, and generate new categories for any goals that do not fit.

Performance Measures vs. Performance Goals vs. Levels of Performance

First, here are some definitions:

• A performance measure is a set of defined outcome-based conditions or response times that project personnel (Owner Agency and contractor) use to evaluate the success of the contractor.
• A level of performance is a defined condition or response time.
• A performance goal is the minimum acceptable level of performance for a given performance measure.

The Owner Agency will need to decide whether it will use pass/fail performance goals or multi-level performance measures.

Pass/fail goals are easier to define, but do not provide as much information on performance. For example, did the contractor just fail the goal or did they badly fail? Did they just pass another goal, or did they truly go the extra mile and exceed the requirement?

If the Owner Agency uses pass/fail performance goals, then the "performance measure" and the "performance goal" are the same thing. If the Owner Agency uses multi-level performance measures, then the measure will include multiple levels of performance. A level of performance is a defined condition or response time. In the multi-level case, the performance goal is the level of performance that is considered a "pass."

If the Owner Agency chooses the multi-level route, the "goal" remains the passing level, and the Owner Agency defines "levels of performance" for the other levels. The Owner Agency will want to make sure that no matter what the outcome, it will fall in to one of the levels of performance. Thus the highest level and lowest level should only have one threshold boundary. The Owner Agency should consider defining a "rejection" level to specify at what level the work is rejected and must be re-done.

The set of "levels of performance" including the "goal" makes up the performance measure. If the Owner Agency chooses the pass/fail route, then they can just stick with the goal.

Establishing the Baseline and Testing the Goals

A large part of determining whether a draft performance goal is achievable - both from the Owner Agency's perspective as well as from the contractor's perspective - is to compare it to the baseline conditions. For example, if the site is experiencing heavy congestion now with no construction in place, it is likely unreasonable to set a goal of no congestion during the construction period. The Owner Agency should compare each goal against the baseline conditions and results achieved in other innovative projects. If the goal seems unreasonable, the Owner Agency should adjust it.

Defining/determining the baseline conditions can involve using available historical information such as crash rates at the site or field data collection to determine current traffic patterns. This will cost money and time, but the information has multiple uses. By providing this baseline information to the contractors as part of the solicitation package, it helps them to better establish their risk level, which helps to get a more accurate price. The Owner Agency will also need this baseline information during the contract to measure performance for some of the defined goals. A number of goals in the sample materials provided in the following section compare performance to "pre-construction" conditions at the site.

Lessons Learned from Real-World Performance Contracts

The following lessons have been learned in real-world performance contracts:

• The goals MUST be something that the contractor can influence.
• Take advantage of the experience of others - do not start from scratch
• Identify the PM/PE/COTR and involve him/her throughout the entire process
• Performance goals form the basis of the contract
• Sources of performance measures and goals include:
  • Agency Goals
  • Common industry standards
  • Research
  • Measures/standards from other agencies/contracts
  • Subject matter experts
  • Brainstorming/working sessions
• Make sure that you cover everything
• Focus on what to achieve, not how to achieve it
• Include time-response goals as appropriate
• 5-level measures are more informative than pass/fail measures
• Consider the possibility of using surrogate measures if the primary measure is not practical to measure in a timely manner
• Test the goals in the field before advertising your RFP/IFB
• When nearing completion of the draft set of measures/goals, it is important to sit back and think, "What have we missed?"
• Defining performance measures/goals is an iterative process
• Getting reviews/approval/buy-in from the offices that will be impacted is very important
• Clearly define what it means to meet the performance goals
• When developing performance goals, it is important to consider how they will be measured/ evaluated and define this in the solicitation. These issues are primarily covered under measurement methodology.
  • Who will collect and analyze the information?
  • Is specialized equipment required?
  • How often will performance be evaluated?
  • Will the evaluation be subjective or objective?
  • Who pays?
  • What happens if a performance goal is not met?
• Total Quality Management and the Environmental Management System are both aimed at creating measures for performance. Expertise and experience in these arenas could help in performance contracting.

Sample Performance Measures

The following table provides a sample menu of performance measures developed for use on construction projects. This set is largely focused around the overall Highways for LIFE goal set. Each agency will need to develop a set of goals that suits their specific project. This sample menu will provide a head start, and will help to accelerate the process.

The sample performance measures are categorized by benefit category (i.e., safety, construction congestion, quality, time, cost savings, customer focus/user satisfaction, environmental, and innovation), and each performance measure has 5 levels of performance. The "Good" or "4" level is the performance goal, or "pass," which denotes the acceptable level of performance.

For general construction projects, and for HfL projects using performance contracting, owner agencies are encouraged to use this sample set of performance measures as a starting menu, and to develop their project performance measures and goals using a subset of these sample measures and goals. All numbers/values presented are samples and must be set to appropriate levels for the locality and the specific project. Owner agencies should also pay close attention to the "Notes" column for each performance measure for important information on each measure.

This sample set of performance measure and goals should greatly accelerate the process of performance measure and goal definition, because the Owner Agency will not need to start from scratch.

One item to stress is that whatever goals are chosen, they must be under the influence of the contractor for that specific contract.

Table 1. Sample Performance Measure/Goal Menu

Sample Performance Goal Materials - MDOT M-115 Pilot Project

MDOT developed the following performance goals for the M-115 pilot project. As part of the RFP/proposal process, MDOT allowed contractors to propose more aggressive goals than MDOT's specified baseline goals. If the contractor proposed more aggressive goals, they received additional points in the best value award, but the proposed goal became the baseline for applying incentives and disincentives.

1. Open to Traffic: M-115 travel lanes are fully open to traffic (no flag control/lane closures or signal operations) on or before the set BASELINE date of August 2, 2008.

Incentive/Disincentive: User delay costs for average weekday with flag control = $7,000.

• Incentive for Open to Traffic prior to the BASELINE date = $7000/calender day.
• Disincentive for Open to Traffic after the BASELINE date = $7000/ calendar day.
• Maximum Incentive = $98,000 (14 calendar days).
• Maximum Disincentive = Unlimited.

Example: The set BASELINE is August 2, 2008. Contractor proposes an open to traffic date of July 2, 2008, which will now becomes the BASELINE. Open to Traffic incentives and disincentives will be base on the BASELINE date of July 2, 2008.

2. Construction & Cleanup Completion: All construction and cleanup roadway and bridges are complete on or before the set BASELINE of 15 calendar days after the actual Open to Traffic Date.

Incentive/Disincentive: $2,650/calendar day

• Incentive for construction & cleanup before the BASELINE number of calendar days = $2,650/calendar day.
• Disincentive for construction & cleanup after the BASELINE number of calendar days = $2,650/calendar day.
• Maximum Incentive = $37,100.
• Maximum disincentive = Unlimited.

Example: The set BASELINE is 15 calendar days after the actual Open to Traffic Date. If the Contractor proposes a construction and cleanup complete of 10 calendar days, the proposed 10 calendar days will now becomes the BASELINE. Construction and Cleanup incentives and disincentives will be based off the BASELINE of 10 calendar days

• Initial Pavement Acceptance: The Pavement Acceptance shall be as specified in the Special Provisions included in the proposal package.
• Pavement Performance Warranty: The set BASELINE warranty period is five years. This allows the Michigan Department of Transportation (MDOT) to waive acceptance testing which reduces inspection requirements. Contractors are encouraged to offer a longer warranty period. This will provide value to the Contractor's proposal and will be accounted for in the determination of the best value aspect of the bid proposal.
• Ride Quality: The ride quality is set with an incentive and there is no disincentive. The Contractor will be required to meet the minimum ride quality specifications. Ride Quality Index (RQI) units are inches per mile

Incentive per ½ Mile Direction:

• Ride quality index between 20 and 30 inches per mile - $2,500
• Ride quality index between 0 and 20 inches per mile - $5,000

Bonus Incentive Entire Project:

• Ride quality index below 30 inches per mile - $25,000

No disincentives apply; the Ride Quality Index must be 30 inches per mile or less

4. Worker Safety During Construction: Worker injury rate (total recordable case rate) less than the rate of 4.0 based on the OSHA 300 rate is the goal for this project.

Method of Measurement: Form OSHA 300A

• Disincentive - $5000 if actual rate is higher than the goal for the duration of the project.
• Incentive - $5000 if actual rate is less than the goal for the duration of project.

5. Work Zone Crashes: Maintain the total pre-construction crash rate of no more then 1.0 crash per month on this 5.56 mile section of roadway for the duration of the project.

Method of Measurement: Transportation Management System (TMS) crash data. The crash data pulled from TMS is from the state-wide data base of actual police crash reports. The data used for measurement will be from actual construction start date to project Final Acceptance date. All crashes during this period will be used, regardless whether there is active construction or not.

• Incentive = $20,000 if equal to or less than 1.0 crash per month.
• Disincentive = $5,000 if equal to or more than 2.0 crashes per month.

6. Motorist Delay: Stage operations to minimize motorist delay. No vehicle shall be delayed due to Contractor's operations more than 10 minutes beyond its normal travel time. Change work operations as needed, to maintain delays below this maximum.

Method of Measurement: On-site total travel time measurements. The random on-site delay measurements will be taken four times per week, twice during the weekdays (Monday - Thursday) and twice on the weekend (Friday - Sunday). Each measurement will include both directions of travel. The measurement for the direction with the highest delay will be used for determining incentive / disincentive. The random on-site measurement will occur between 10:00 am - 1:00 pm and 3:00 pm - 6:00 pm, with a +/- 30 minute variance. Normal travel time @ 55 MPH for 11 miles = 12 minutes

Incentive / Disincentive per Measurement:

Maximum Incentive = $50,000

Bonus Overall Incentive: If there are no more than 3 measured occurrences exceeding 10 minutes and less than or equal to 15 minutes delay for the duration of the project, the Contractor will be eligible for the Bonus Overall Incentive. Any one measurement exceeding 15 minutes will cause the Bonus Overall Incentive to not apply.

Bonus Overall Incentive = $50,000" ¹

¹ Michigan Department of Transportation RFP -84169, Highways for LIFE Project. [Return to note 1. ]