Accuracy is established after the estimate has been completed and a Monte Carlo risk analysis has been performed. If, however, we focus on quality first, improved accuracy should naturally follow.
The recommended industry standard within the oil, gas and chemical industry was established by the AACE International Recommended Practice 18R-97. This practice establishes usage, preparation methodology and accuracy as the classes of estimate and which class is appropriate for a given phase of a project. Having this industry standard is all well and good, but it doesn’t address quality. So, what can we as estimating professionals do to improve the quality of the various classes of estimates?
Improving the quality of estimates starts with improving everything that directly affects them: well-trained estimators, adherence to standard operating procedures and best practices, inputs, execution strategies, project schedules, unit costs and identified project risks.
Having well-trained estimators tops the list when striving to improve the quality of estimates. This extends beyond being able to simply use a piece of software. Without an understanding of how a project will be executed or the processes required to execute the work, an inexperienced estimator cannot produce a quality estimate. As such, time and money must be invested in training the estimator to develop the required knowledge. This training must include time working on projects in the field. Field experience provides the required fundamentals—no software package can account for that.
This is an area where seemingly minor errors can have a major impact on an estimate. Some of the most common errors include the following:
- Using incorrect design parameters with programs such as Aspen Capital Cost Estimator
- Inputting the wrong unit rate for unit cost for a reimbursable unit rate contract
- A misplaced decimal place when entering a quantity, unit manhours or unit cost
- Omitting a unit cost or manhour all together
Data entry errors can have impacts exceeding a million dollars. Training the estimators so they understand how to adjust for differing design parameters, units of measure or unit rates is fundamental to using a program such as Capital Cost estimator. It is not uncommon for the information on a data sheet to show pressure in Barg or kg/cm2 when the program requires pressure in kilo-Pascals (KPa). Entering the pressure without first converting to the proper unit can have a major impact on the cost.
Providing the estimators enough time to prepare the estimate is essential to eliminating input errors.
Providing the estimators enough time to prepare the estimate is essential to eliminating input errors. All too often engineering and procurement schedules slip but estimating must hold to the original schedule. This abbreviated schedule dramatically increases the likelihood of data entry errors. Extended periods of overtime lead to fatigue-induced inefficiencies and mistakes. In addition, out-of-sequence work creates situations where updates are not made or are made incorrectly. Input errors typically caught and corrected during the project team review cycle are missed due to the condensed schedule. These errors are likely to be missed during the corporate review cycle as well. With corporate reviews focusing on higher level issues such as alignment between the estimate and the underlying technical documents, procurement strategies, execution basis and compliance with bid conditions, issues like input errors are assumed resolved by this point.
For an estimate to be valid, it must reflect the scope of work and the execution strategy. I recently assured an estimate for the replacement of a process unit in a refinery. The estimate was prepared for the correct scope of work but missed the execution strategy. It was prepared for a stick-built facility when it was to be executed as a modular project to reduce the required work hours at site. This disconnect in the execution caused a complete recycle of the estimate which delayed approvals by four months. A good understanding of the execution strategy is important, as it will directly impact costs.
The project schedule allows the estimator to properly identify the cost of the time-related items and forward escalation costs. A well-developed schedule is a joint effort by planning and estimating. Planning requires the final estimated quantities to validate the durations and work-hours to establish manpower histograms. The histograms can be used to identify congestion issues, which may affect labor productivity and the costs. Once the schedule has been validated, the estimator can better estimate the time-sensitive costs, including escalation and general conditions.
Everyone understands the importance of well-defined unit costs. In theory, if estimators use better quality unit rates they will generate better-quality estimates. This is true if the unit costs are applied properly. The mis-application of unit costs can be difficult to spot during reviews and can go unnoticed.
We must first recognize that unit costs not only include the cost of mechanical equipment, bulk materials, labor rates which comprise direct costs, but also other items such as foreign exchange rates, freight, import duties, taxes, insurance and letters of credit that are mainly indirect costs. By recognizing this, we can better understand where and how unit costs might be incorrectly applied.
For junior-level estimators the most common errors will be the mis-application of a unit rate to a cost item, omitting a unit cost or incorrectly converting the cost of an item in a foreign currency to the equivalent bid currency. These input errors can be eliminated through proper training. For more senior estimators, the proper application of a unit cost can be more complex. They must understand how the unit costs are to be applied. In the case of taxes and duties, this is not always easy as there will often be items that are exempt from these charges. The mis-application can either leave a contractor exposed or cost them the bid due to overestimating the value. For an owner, it may lead to unnecessary budget overruns. As such, the estimator must have the proper knowledge to calculate the cost accurately or know who to go to for the proper calculation. Costs such as insurances and letters of credit may be estimated as a percentage of total contract value and not as a multiplier of all the previous costs. This is a subtlety that can result in a cost being underestimated. An example of the miscalculation of taxes and the impact to a project’s profitability occurred when a contractor failed to take into consideration the local income taxes for expatriate workers in Canada. The cost to the project turned it from one that should have been profitable to one that was going to lose money from the moment the contract was signed.
Project risks are separate from the uncertainty ranges set in the Monte Carlo analysis to establish project contingency. These risks are discrete events that may occur during execution. The project team must determine how to manage the risks associated with any project. If a risk is to be mitigated, the mitigation costs must be included in the base estimate. The residual or remaining risk (post-mitigation) then becomes part of the Monte Carlo risk analysis. Identifying the cost of the mitigation actions in the estimate is ideal. If the mitigation actions are not accepted when the inevitable cost reduction exercises begin, the associated value can be removed, and the Monte Carlo analysis can be adjusted to reflect the increased risk to the project.
If we are to improve the quality of estimates, we must first focus on improving the skill sets of the estimators and invest in their training. As their knowledge of what they are doing and why they are doing it improves other focus areas should also improve. Well-trained estimators are an integral part of the project team, and equipping estimators with the knowledge and skills to work with the other disciplines on a project ensures the estimate reflects the scope of work and execution strategy.