Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
32 CHAPTER 3 CRITICAL REVIEW OF THE STATE OF PRACTICE The main goal of Task 2 is to review current practices in the area of cost estimation practice and cost estimation management. Numerous unique or innovative approaches to cost estimation practice and cost estimation management are described. A discussion of how current cost estimation practice and cost estimation management approaches do and do not address the identified potential root causes of project cost escalation are also covered in this chapter. Finally, important issues and more specifically deficiencies in current practice are identified. Figure 3.1 summarizes the inputs and outputs of this task. TASK 2 (CHAPTER 3) FACTORS CAUSING COST ESCALATION UNIQUE PRACTICE APPROACHES GENERAL DEFICIENCIES IN PRACTICE GENERAL OVERVIEW OF CURRENT PRACTICE CRITICAL REVIEW Figure 3.1. Critical Review Inputs and Outputs METHODOLOGY The research methodology followed for Task 2 builds upon Task 1 findings. The data and information collected through the interviews, combined with results of an exhaustive literature review, assisted the team in identifying reasons for cost growth and estimate inaccuracies. A detailed data analysis meeting took place in College Station, Texas, where the team performed a critical examination of the information collected for the state of practice. The Task 2 interview data was analyzed according to project development phases. The research team reviewed the interview data and identified unique and/or innovative approaches that will aid the Departments of Transportation in overcoming factors that cause project cost escalation, as identified in Table 2.2. The research team used literature, interviews, and experience to link the better DOT approaches to cost escalation factors these approaches would address. However, the research team did not have sufficient information about each approach presented in this chapter to measure the effectiveness of the approaches in addressing cost escalation. Further, the team did not gather effectiveness information directly during the interview process. Our intent is to address these issues in Phase II of the research, and more specifically, through Taskâs 6 and 7. In general, all those factors causing project cost escalation, as noted in the literature and discussed in this report, receive some attention by DOTs; however, not every DOT addresses all of these factors in their entirety. Therefore, the cost estimation practice and cost estimation management techniques described in this chapter are a compilation of the better practices from many transportation agencies, transportation consultants, and a single private non-transportation company.
33 A discussion of current practices follows. This discussion is organized into categories of: 1) cost estimation practice; and 2) cost estimation management. These categories are further organized into the estimates that correspond to the major project development phases. Each of the practices listed in Table 3.1 below is discussed. Once the practices are described, the research team noted potential deficiencies in current practice. These deficiencies are discussed under âSummary of Important Issues,â the last section of this chapter. These unique practice approaches and the general deficiencies in practice are the basis for the strategies, methods, and tools described in Chapter 4. Table 3.1. Current Cost Estimation Practice and Cost Estimation Management Project Development Phase Cost Estimation Practice Cost Estimation Management Planning Estimates ⢠Applying Cost-per-Mile Factors Using Typical Sections ⢠Applying Cost-per-Mile Factors Using Similar Projects ⢠Order of Magnitude Estimates ⢠Add-on elements ⢠Communication ⢠Conceptual Estimating Software ⢠Red Flag Items ⢠Recognition of Project Complexity Programming and Preliminary Design Estimates ⢠Identifying Major Cost Items ⢠Conceptual and Parametric Estimating ⢠Volumetric Estimating ⢠Risk Analysis ⢠Add-on elements ⢠Estimate Reviews ⢠Scoping Documents ⢠Communication ⢠Public Involvement ⢠Conceptual/Parametric Estimating Software ⢠Definitive Management Plan ⢠Risk Charter ⢠Estimating Checklist ⢠Design Value Engineering ⢠Design to Cost ⢠Management of ROW, Utilities and Environ. Issues ⢠State Estimating Departments ⢠Cradle to Grave Estimators ⢠Year of Construction Costs ⢠Scope Change Form ⢠Cost Containment Table ⢠Gated Process ⢠Create Project Baseline ⢠Estimate Manual ⢠Estimator Training ⢠Estimation Scorecard Final Design Estimates ⢠Estimating Software ⢠Historical Bid Price Databases ⢠Estimation and Management Software ⢠Estimate Review
34 When interviewing DOTs, the distinction between programming and advanced planning/preliminary design was not always clear. Further, the estimation procedures employed by the DOTs were in these two phases were similar. As a result, we have combined these two phases in the discussion to follow and when listing preliminary methods and tools in the next chapter. COST ESTIMATON PRACTICE Planning Estimates The long-range planning estimate is usually the first estimate produced for an identified need, that is, a future project. When the identified need is added as a project to the DOTsâ long-range plan, the estimated cost is an important criteria often used to prioritize different needs within the transportation program. Additionally, the purpose of this estimate is to determine funding levels for long range plans. As described in the state of practice, the typical method used to prepare planning estimates is historical lane-mile cost averages. Applying Cost-per-Mile Factors Using Typical Sections Cost-per-Mile Estimation Handbook One unique approach to applying cost per mile factors is developing typical project sections (e.g., pavements) that correspond with the lane-mile cost factors. Using this approach, one DOT created an estimation handbook, which has sketches of typical project sections that are used to generate the conceptual estimate. At the planning stage, the pavement thickness, materials, and lane widths are typical numbers. Depending on the projectâs standard characteristics, the estimator chooses the corresponding project typical from the handbook. Then, the estimator selects the appropriate cost chart that best fits the anticipated project structure. Cost is still in dollars per a lane mile but it reflects a proposed typical section, a typical structural section that is identified early in project development. The typical sketches also aid the estimator in deciding on the additional project elements that will be required. The base construction cost, and therefore, the preliminary engineering, civil engineering, inspection, and right-of-way costs are added to this lane mile cost. The right-of-way (ROW) is factored into the estimate as a percentage of the estimated construction cost, and the engineering costs are based on historical ratios of engineering to construction cost. The engineering cost includes preliminary engineering, construction engineering inspection, right-of-way support, and related overhead costs. The factors in the DOT handbook represent present day costs, which must be inflated to the projectâs midpoint of construction. The planning manual has inflation factors that are applied to the planning estimates. The calculated elements are summed to arrive at the long-range planning estimateâs total amount. This estimation method provides the DOT with a consistent and transparent approach to costing projects. Consistency of approach continues as the project is further developed because the DOT uses an estimation methodology that builds upon the lane mile typical section at each project development phase. The difference between the estimates in each phase is the incorporated level of project detail. Furthermore, the estimate is documented by the systematic preparation of narratives. The approach also has standard project cost components that must be considered for inclusion in the estimate; this helps the estimators avoid the problem of cost item omission.
35 Cost-per-Mile Spreadsheet Templates Two DOTs reported using lane-mile cost factors with typical sections for their planning estimates, but their methods were not as consistently used within the DOT as the procedure previously described. One of the DOTs uses three Excel spreadsheet templates specifically for its central, northern, and southern regions. The templates categorize typical projects into rural or urban location, and into new or widening projects. The number of roadway travel lanes and the median type is used to further define each typical section. The Excel spreadsheet templates have columns associated with costs for grading and drainage, base aggregate and pavement, lump sum items (i.e. pavement markings and signs), miscellaneous items, engineering and contingency, total project cost, and total cost per mile. The length of the proposed project is entered into the template, and costs for each typical section listed are calculated. This template provides the DOT with different design alternatives along with an estimate for each design so that designs can be compared. Similarly, another DOT has a cost sheet that lists similar project types and associated cost per mile factors. The cost sheet separates projects into rural and urban with project types listed by the number of roadway travel lanes. From the cost sheet, the estimator chooses the thickness of the pavement and the median type. The cost sheet also refines cost numbers based on work type, reconstruction or new construction. Furthermore, the sheet provides information for estimating the cost of miscellaneous improvements such as signaling. Percentages are used to estimate right-of-way and utility cost. This DOT is in the process of refining their estimation software to include the computerization of planning estimate preparation. Applying Cost-per-Mile Factors Using Similar Projects Several DOTs use information from similar projects that have been fully designed to generate cost per mile factors for long-range planning estimates. One transportation agency identifies similar type projects within the state that are in the programming phase and uses the current average cost per mile estimates from those projects to prepare the conceptual estimates for its planning phase projects. The cost per mile cost data could be obtained from a single programmed project or from a number of similar programmed projects. The planning engineers in the respective districts provide the estimators with the current cost per mile estimate for the programmed projects. Thus, the conceptual estimates reflect all project costs elements, including costs for design, utilities, construction, and right-of-way. If the project includes structures, the estimator tries to segregate and remove the structure cost in the programming phase estimates and then estimates the current projectâs structures separately. Other DOTs develop lane-mile factors in a similar manner as the one describe, but they use costs for projects that have already been let instead of projects still in the programming phase. Scoping Document One DOT creates its long-range planning estimates using costs from similar projects, but they also use a scoping document in creating the estimate. The scoping document separates the project costs into five categories related to general roadwork: pavement structural section, roadwork, drainage, specialty items, and traffic items. These major elements are estimated using historical bid averages. Minor items, mobilization, and roadway additions are estimated as percentages of the roadway items. The Departmentâs structure and right-of-way divisions are
36 responsible for generating estimates for their project elements. Most projects are informally compared to similar existing projects to check for consistency. Order of Magnitude Estimation Early in project development, estimates are often done very quickly with only limited project definition, and expected cost is communicated as an order of magnitude estimate. One DOT expresses the long-range planning estimate as an order of magnitude estimate so those who see the estimate are aware of the limited scope definition that was used to prepare the estimate. These estimates should only be used for the very initial feasibility studies. Order of magnitude estimates are far from exact, and only represent an indication as to a degree of expected project cost. The plus or minus 40% confidence range typically associated with these estimates reflects the lack of definite project information (Merrow, Phillips, and Myers 1981), however one DOT reported a 45% confidence range. Add-On Elements All DOTs incorporate in one manner or another affects of âAdd-onâ elements have on project cost. These âAdd-onâ elements often result from local government concerns, environmental issues, and externally imposed requirements. During the long-range planning phase, these issues are added into the estimate as a percentage value based on the total project cost. The percentage either is identified as a separate cost item or is incorporated into other items such as miscellaneous, preliminary engineering, or contingency. Programming and Advanced Planning/Preliminary Design Estimates As a project moves into the programming stage of project development, the techniques used to create the cost estimate changes to reflect the availability of additional project information. During the critical analysis of the state of practice, the research team discovered that the programming and the preliminary design phases possess many similarities. For many DOTs, programming and preliminary design overlap one another, and the programming estimate is often considered a milestone established within the preliminary design phase. Because of the similarities between the two phases, they were combined for the critical review. Identifying Major Cost Items When a project is in the programming and preliminary design phase, more information about the project scope is developed. Therefore, the estimates created for the project are more specific than the earlier estimates. Many DOTs recognize the fact that about 80% of the project cost is in about 20% of the project elements. As a result, these DOTs focus on the high cost items while generating an estimate. The DOTs that identify costs for major items use a spreadsheet or in- house software to calculate the total estimate. One DOTâs list of major items included surfacing, safety items, structures, and grate and drainage. Another DOTâs major items are: excavation, embankment, bituminous pavements, Portland cement concrete pavements, drainage, curbs and gutters, structural concrete, structural steel, and guard rail. In both cases, these large cost elements are estimated using historical unit costs or cost-based estimation procedures. Once the major items are estimated, the smaller items, such as traffic control, signing, and striping, are included as percentages or by lane-mile factors similar to those used in planning estimates. By applying this estimation approach, the DOTs are considering the projectâs major cost drivers and
37 the projectâs complexity. The DOT focuses on the major cost drivers and attempts to develop a precise estimate for those items. Although the minor items are not estimated at the same level of detail as the major items, they are identified and incorporated into the estimate by methods that are more global. Conceptual and Parametric Estimation In-House Estimation Software (long-range estimation) For the programming and preliminary design phase, a few DOTs are using computer software to develop conceptual and parametric estimates. For one DOT, the information in their in-house estimation software is recorded in a handbook that is used for the conceptual planning estimate. When the project reaches the programming stage, the DOTâs project development group creates different alternatives and then chooses the one that best meets the projectâs needs. Then they use their in-house estimation program to produce the program estimate. Each section of the project can be described by a different typical sketch. The estimator starts with a preloaded typical and then adjusts it according to the site conditions and project location. The location can be specified by county, market area, or general statewide information can be used. At the programming stage, the estimate becomes more project specific. The DOT tries to perform parametric estimation by identifying the major cost items, such as sound walls, structures, retaining walls, and required clearing. The estimator should visit the project site and decide which work items need to be included in the estimate to reflect specific site conditions. This same program is used to create preliminary design estimates. Scope of Work Estimation Software The key to another DOTâs estimation software is a complete scope of work. Therefore, the estimate prepared for the programming and preliminary design phases are scope feature driven. The estimation system includes lane-mile cost for nine geometric conditions, which are based on the functional classification of the roadway and the terrain. The user must specify in the system when the project will be constructed, and the cost is adjusted according to the entered date. In addition to the lane mile geometric conditions, the cost for the other project items such as structures, demolition of existing structures median barrier, curb and gutter, signals, and crossovers must be estimated and added independently by the estimator. The project manager or the estimator can also add features and costs that were developed outside the system and input those costs into the estimate. An example of such an additive would be the additional costs for extensive phasing or for productivity impacts for projects in an urban environment. The remaining cost elements of a project such as for design, construction engineering, inspection, and right-of-way are drawn from detailed cost models. Design costs are extracted from a curve of historical construction cost versus the value of road design and a separate curve is used for bridge design. Construction Engineering and Inspection (CEI) costs come from a curve based on historical close out cost information. These curves are built into the project cost estimation system. Although the system has right-of-way (ROW) models that are based on the amount of ROW and the current land use, the estimator has the option to apply a cost derived independent of the estimation system. Once the engineering drawings are complete and all quantities are known, the user can chose the Trnsâ¢port section of the estimation system to create
38 an estimate. This DOT allows the public access to the systemâs project information creating transparency for the DOT. The openness helps prevent tendencies to create a biased estimate. AASHTOâs Trnsâ¢port Software Computers and estimation software enhance the ability of engineers to manage large data sets that can be used in developing estimates for all types of projects. In the case of DOTs the most widely used estimation software is Estimator⢠by InfoTech. Estimator is a module of Trnsâ¢port. Trnsxâ¢port is owned by Info Tech, Inc. and fully licensed by AASHTO. Using this software DOTs can prepare parametric or item level project cost estimates. Parametric estimates are based on project work types and their major cost drivers. Item level estimates can be derived either from bid histories or by using cost-based estimation techniques. Cost-based estimates are based on an assumed productivity and the direct cost of material, equipment and labor. A survey of DOTs conducted in the fall of 2002 found that the Trnsâ¢port Estimator module was being used by 22 DOTs at that time. Historic bid price databases can be created using the BAMS/DDS module of Trnsâ¢port. BAMS/DDS is the Decision Support System module of the construction contract information historical database. Another commercially available system that is used by several DOTs is âBid Tabsâ by OMAN systems. It is used either as a stand-alone or in conjunction with âTrnsâ¢portâ by seven DOTs (Schexnayder et. al. 2003). Department in-house and AASHTO estimation software are tools that assist the DOTs in developing their project estimates. The estimation programs with preloaded templates help the DOT project teams define the project scope, cost, and schedule. The software provides a means to track project development, and it can assist in project review. Due to software flexibility, the estimator can adjust unit costs or percentages according to the projectâs complexity. Estimation software also permits the easy inclusion of additional items that are unique to a particular project. Volumetric Estimation Another procedure used to create the programming and preliminary design estimate is a volumetric method based upon the pavement component of a project. For this procedure, a length-width-depth (LWD) template has been developed by the DOT for generating planning estimates. Basic project information such as scope of work and the control section are entered into the template. Then the LWD factors for all the roadway items are determined. After that, an LWD cost multiplier is selected for a table and entered into the multiplier box on the template. The estimator must generate the costs for the other project design elements and enter them into the template. The template sums the individual roadway item costs, totals the cost column, and advances that cost to the project total box. The last step is completing a âProject Scope Summary Formâ for the estimate. The Length-Width-Depth cost accounts for all costs associated with building the roadway; it represents the ânormalâ cost for major items of construction, such as: mobilization, removals and salvage, grading, aggregates, paving and approach panels, by-pass and temporary construction, drainage, concrete items, traffic control, turf/erosion, and miscellaneous. The estimator will collect all the LWD information and break the information up into two portions. The LWD portion is an accumulation of all the roadway parts, and it is used to create a project cost multiplier related to the unit volume consisting of pavement, shoulder, or rampâs length, width,
39 and depth. The project LWD factor is the sum of the volumes (LWD factors) of all the roadway items in the project. The depth of pavement does not include the aggregate base or sub-grade. Depths selected by the departmentâs Estimate Coordinators are based on historical data and/or as project scopes dictate. The project LWD factor (volume) is multiplied by a LWD cost multiplier that has been developed through historical data and represents different projects with similar type and scope. The DOT created a menu of project types along with a cost multiplier for each type. The department also has indicators to follow such as a cost per a square foot of pavement or cost per a lane mile of pavement to check the LWD estimate for reasonableness. Five specific cost items are not included in the LWD factor roadway cost estimate and must be computed separately. Those five cost items are bridges, signals, noise and retaining walls, traffic management systems, and other abnormal construction items. Other cost items that must be added to the LWD cost are: engineering, right-of-way, and relocation of utilities. A percentage additive item is used to account for project development costs, including engineering, design, and construction costs. About 20% of the project cost is typically used for this item. For the right-of-way (ROW) cost, the DOT expects that the engineers will layout the project and develop the cost. The engineer assumes a distance from the edge of pavement and that sets the right-of- way limit. A parcel database from the stateâs geographic information system allows the estimators to determine which parcels are impacted by the assumed right-of-way. At this point in project development any impacted parcel is assumed to be a total take. The County Assessor provides information on the assessed market value for the impacted parcels. A multiplier, specific to the corresponding county, is applied to the parcel value. The cost of the parcels is totaled to obtain the right-of-way cost estimate. Once all of these project elements have been calculated, they are added together to provide the total planning estimate value. Risk Analysis As described in Chapter 2, State of Practice, each DOT addresses contingency differently, with 1) a fixed percentage, 2) a sliding scale, or 3) a structural/formal analysis being the most common approaches. Although most interviewed DOTs factor contingency into their estimates, only one DOT performs a detailed risk analysis, using a tool developed by the department. When this DOT creates an estimate, they remove all contingencies from the line items. Then, the DOT develops a base cost and schedule that represents performance of the project according to the plan. After that, cost risks, schedule risks, and opportunities are identified and evaluated. The DOT combines the base cost and the risk/opportunity assessment and then applies critical path methodology and Monte Carlo simulation to generate ranges for expected project cost and schedule. The methodology also generates related probabilities for the predicted cost and schedule ranges. Through this risk analysis tool, the DOT has created a method for applying contingency factors that are based on an in depth analysis of possible events and the probability of the eventâs occurrence. By performing this analysis the DOT recognizes potential project problems early in project development process and this enables the Department to respond proactively to the identified events. Add-On Elements During the programming and preliminary design phases of project development, every department considers âAdd-onâ elements while developing the project estimate, but at this point in project development âAdd-onâ elements are considered separate from direct project line items.
40 Many DOTs have established environmental assessment as a project milestone. Therefore, an estimator must consider any environmental or cultural issues that can affect the cost of the project. If the environmental assessment is not complete, then one DOT has a policy of not assigning funds to the project. Other DOTs perform these âAdd-onâ elements evaluations during their internal estimate reviews. During the reviews, they address issues such as environmental mitigation, public involvement, and context sensitive design issues that might hinder the advancement of the project. Estimate Reviews For reviewing a programming or preliminary design estimate, one DOT conducts peer reviews. The project manager and the design team review the design and comment on any discrepancies or problems. The designers of the specialty items such as retaining walls and structures make certain their features are accurately represented in the design and estimate. By reviewing the estimates, the DOTs can detect possible errors or omissions. DOTs also use reviews to identify discrepancies in the estimate that are the result of bias that lead to underestimation of project cost. Another DOT stressed the importance of gathering the individuals responsible for all the different aspects of the project such as right-of-way, structures, and surveying so that their input could be utilized to develop a realistic estimate. The DOT also explained that involving all disciplines early in the project development process is important to the projectâs final outcome. Final Design Estimates Once a project has entered the final design phase, the projectâs scope should be completely developed, and therefore, all project elements can be estimated with precision. This higher level of project knowledge enables DOTs to create a detailed estimate. Furthermore, estimation and management software is typically used to assist the DOT in producing the final design or engineerâs estimate. Although previous estimates are prepared by the DOTsâ district or regional office, the final or engineerâs estimate is typically completed by the DOTâs central office. When the projectâs design is ready for advertisement, it is sent to the central office, and a detailed estimate is prepared by the headquartersâ staff. The final estimate is produced using the bid items and plan quantities derived from the completed plans and specifications. Then, applicable historical unit cost data that has been adjusted to reflect current year costs is applied to the quantities. A few DOTs generate the engineerâs estimate within the district or region and then sent it to the central office for review prior to the letting. Estimation Software Often DOTs use estimation software to calculate the engineerâs estimate. The software is either a program that has been developed within the department or the Estimator module from AASHTOâs Trnsâ¢port software. A few DOTs use a combination of their in-house software and the AASHTO programs. The DOTs that have AASHTOâs Trnsâ¢port use one or several different modules of the software, such as the Cost Estimation System (CES), the Proposal and Estimates System (PES), or the Estimator module. The Cost Estimation System enables the user to prepare parametric and cost-based estimates. The CES module has the ability to store historical labor, equipment, material, and crew data. Detailed project information can also be entered into the program. If a DOT uses the Proposal and Estimates System, they can enter project data into the
41 program and prepare conceptual to detailed estimates. Within PES, the DOT can use multiple funding units and differing percentages for engineering and contingency. AASHTOâs Estimator module allows the user to use several different estimation methods such as estimates based on historical bid data, historical cost data, reference tables, or a collection of price derivations. All the data used to generate an estimate such as crew wages, equipment and material costs, production rates, and historical cost data is stored in Estimator. Historical Bid Price Databases Along with estimation software, DOTs have extensive databases of their accumulated historical bid data. All of the possible items that would be used in a project are set in these databases, and each item is tied to a specific specification. A staff unit at the DOTâs headquarters often manages the database, with the districts and regions having on line access to the information. Departments vary as to the period of time historical data is retained in their databases and how far back price data should be considered to determine average prices used in estimates. Typical look back periods are 1 year, 18 months, or two years for use in averages. Nine DOTs retain data for as long as records exist (Schexnayder et. al. 2003). Estimators can examine and use this data for items that are not frequently encountered or items that have seasonal price swings as an averaging of data obscures seasonal pricing. The bid averages shown in the database is calculated several ways: Low bid only - 20 DOTs Low and second bid - 1 DOT Three lowest bids - 15 DOTs All bids (but may exclude single bids that are very high or low) - 11 DOTs All bids except high and low - 2 DOTs Bid analysis to determine a reasonable bid amount for each line item - 1 DOT (Schexnayder et. al. 2003). By the using of different sorting criteria, the line-item cost data can be analyzed under different protocols. The line-item cost data can be sorted by district, county, region, and state. In addition, the data is also categorized by project type, market area, location, and terrain. Within the historical database, the users can view the bid average for a particular item or they can view all the unit prices so the user can select a price that corresponds to their estimated quantities. One DOT database has an item price menu, and the user can view different item criteria, such as a date range, region and county prices, only awarded prices, all bid prices, specifications in English or Metric units, funding, quantity range, similar projects, or contractorâs bid. Finally, bid prices are also used to support in-house programs like the long-range estimation approach. The Trnsâ¢port modules discussed earlier have the ability to store historical bid information and use the data in estimate preparation. The Trnsâ¢port CES program uses historical data and
42 regression models. The regression models take into account specific criteria such as quantity, season, market area, and date. The regression curves help the estimator know how reliable their unit cost is based on the number of criteria it meets. For example, if the regression curves show that 4 out of 6 categories apply to the unit cost used, then the estimator can be certain the unit cost is precise. The Trnsâ¢port BAMS/DSS program also analyzes historical bid information. Within this database, the DOT can view contract and vendor information and analyze the market. The program also assists the DOT in analyzing bids, specifically in searching for unbalanced bidding. BAMS/DSS can assess historical bid prices and estimates, and it can evaluate the difference between the awarded and final costs and quantities of a specific project. The databases allow the DOTs to systematically utilize the large amounts of price information they have collected over time. By using the large databases, estimators can select the most appropriate unit costs for their project enabling them to consider unique project characteristics. If the same database is accessible throughout the state, then the individuals developing the project estimate can apply data that they would otherwise not have available to them. The large databases help prevent estimators from relying on data that is not relevant to a specific project. COST ESTIMATION MANAGEMENT Cost estimation management should occur continuously throughout the project development process. Some efforts are exclusive to a particular stage of development, while others are pervasive throughout the process. The three phases of 1) planning, 2) programming and advanced planning/preliminary design, and 3) final design can require the application of different cost management methods due to the level of project information that is available and the manner in which the estimate must be communicated. From the DOT interviews, it was discovered that a variety of cost estimation management methods are being utilized by DOTs. This section of the report examines the different cost and schedule management methods used by the DOTs. A number of ideas that were found in the non-DOT interviews are also discussed. Planning Phase Communication How estimate precision is communicated is important, particularly during the earliest stages of project development. Communication of Importance - Every project estimate is important as cost is integral to project scope, and together cost and scope drive many of the project teamâs design decisions. Additionally, the estimated costs that are presented to stakeholders outside of the project team can have many positive and negative implications to the project and to the DOT. All team members must understand the importance of cost estimation if costs are to be managed appropriately. Many projects have been misrepresented, in terms of scope and cost, early in their development because of a lack of understanding that estimates must indicate the âtotal costâ of the project, as it is known at the time the estimate is completed. To maintain creditability with stakeholders it is important to âtell the public the truthâ about project cost and identify the precision of estimate values. Communication of Uncertainty - Projects are not well defined in the early stages of their development. Identification and communication of the projectâs early stage uncertainty and the
43 fact that unknowns can impact scope and costs will help in managing project expectations. The unknown elements of a project estimate can be communicated as cost ranges rather than as point values. The wider the range of values obviously the greater the number of unknowns or specific information about the identified unknown. Communication of the uncertainty can aid in communicating the need to better define the unknown elements and perhaps the need to seek ways to mitigate the unknown aspects of the project by engineering and construction approaches. This process of communicating project cost uncertainty can begin in any stage and continue throughout project development. Conceptual Estimation Software When estimation software is used for developing the estimate rather than a simple spreadsheet, the software often has associated cost management tools. The research team was provided examples of both commercially available software and software developed within particular DOTs that had the capability of enhancing project cost management efforts. Examples of this would be estimation software, which required that certain items be entered, checked off, dated, and/or approved before the project development could progress to the next stage of development. Such requirements ensure that all cost related aspects of the project have been properly addressed, limiting later scope changes, and to the practice of inconsistently apply contingency. Many software packages have the capability of highlighting costs that are out of prescribed ranges, thereby prompting the DOT to check the accuracy of an estimate. Estimation software can also be used to track item costs as the project moves from one development phase to another. This feature helps to control total project costs. Currently DOTs do not use any sophisticated estimation software at the planning stage of project development. The majority of DOTs use simple spreadsheets at this project development level. Two DOTs are planning to expand their current in-house software packages used to estimate and manage project costs at the earlier stages of project development. Red Flag Items Items that can potentially impact project cost in a significant way are sometimes identifiedâ¯red flaggedâ¯by DOTs early in the planning process. Many DOTs develop a list of these impacting items, based primarily on engineering judgment. The red flagging of items may not involve any formal risk analysis of these factors. In the case of a DOT having a repair project of an urban interstate bridge that crosses over a commercial rail line and a light rail transit line, the red flag item list may include such things as coordination with railway, maintaining open tracks for the light rail, coordination with the light rail entity, and maintaining sufficient traffic over the bridge for the daily commuter traffic. Recognition of Project Complexity Project complexity should be addressed early in the project development process so that appropriate cost estimation methods are conducted and the project can be properly managed. One DOT has created three tables that describe project complexity. The DOT defines three categories for project complexity: non-complex (minor) projects, moderately complex projects, and most complex (major) projects. For each table, the projects are categorized by project elements: roadway, traffic control, structures, right-of-way, utilities, environmental, and stakeholders. Within each section, the type of projects and criteria are listed. For example, non-complex projects for roadways are maintenance betterment projects, moderately complex projects for
44 roadways are minor roadway relocations, and most complex projects for roadways are new highways. The five other project elements have similar lists. For the stakeholder section, the DOT describes non-complex projects as those that have no public controversy issues. Moderately complex projects moderately involve the public and public officials due to non- controversial project types, and general communication about project progress is required. The most complex projects are controversial and high profile projects, and major coordination among numerous stakeholders is required. The project complexity tables provide a statewide definition of project complexity that ensures projects of similar complexity are subject to the same reviews and attention. These definitions allow for a common language between DOT employees to aid in communication regarding projects. This type of definition insures that estimates reflect appropriate levels of complexity. Programming and Preliminary Design Scoping Documents Many DOTs use âscoping documentsâ early in the project development process to identify and specify critical design elements. These documents create a baseline scope for the project and any changes in the scope are measured against this baseline-scoping document. When used correctly, this document can be an effective cost management method addressing many cost overrun factors including scope changes, project schedule changes, and engineering and construction complexities. Explicitly defining the scope of the project early in the project development phase allows for better scope control and identification of any changes, which may translate to changes in project cost and schedule. One DOT holds a scoping meeting when the project enters the preliminary engineering phase. The meeting brings experts from each phase and discipline together for a field review of the project. The meeting is used to: 1) specify the project limits; 2) identify issues that may affect project elements; 3) agree on the purpose of the project; 4) refine the construction cost estimate; 5) enhance the project schedule; and 6) define the participation of each discipline and establish a contact person. Upon completion of this meeting a specific document must be completed that distills the decisions and information of the meeting. This document is then distributed to various parties. Prior to signing the final plans for either right-of-way or construction another form must be completed stating that the project is within the original scope. If it is not within the original scope, documentation concerning deviations must be provided. Another DOT has a Project Scoping Memorandum that is completed by the project manager and submitted to the Design Technical Support Engineer for review and comment. The memorandum summarizes the important information of the project and certifies the scope is as complete as possible at that point in time. Communication Communication of Uncertainty - The identification and communication of the uncertainty and of project scope and cost unknowns helps in managing project cost in the Programming and Preliminary Design phase just as for the planning phase. As the project moves from programming through preliminary design, the amount of uncertainty in the estimate should diminish. Good cost management techniques communicate specifically how the design process has removed the uncertainty.
45 Communication between Departments within the DOT - Communication between internal DOT departments is imperative throughout project development given the complexity and number of people involved in even the simplest project. One DOT mentioned that communications must be open between all departments, and all departments must be active in the project development process, even during the earliest stage of the development phase. Public Involvement Public involvement is an essential cost management tool. Public involvement in environmental planning and project scoping is commonplace in highway design, but the public is not often sufficiently involved in project cost development. Conducting public workshops with a focus on cost can help to manage and communicate cost impacts more effectively. Communicating the precision of an estimate, prepared at a particular point in project development, is essential to limiting local government concerns and requirements (Construction Industry Institute 1994). Conceptual/Parametric Estimation Software Similar to in the planning phase, estimation software can serve as a comprehensive cost management tool. At this stage, some DOTs still use spreadsheets to prepare estimates and manage costs. However, conceptual/parametric estimation software packages can serve as useful cost management tools in this phase. Some DOTs have developed their own software, and AASHTO Trnsâ¢port software is available. These programs provide management with the ability to compare the original estimates with any future estimate. Definitive Management Plan From the FHWA Major (Mega) Projects Lessons Learned (2003) document, it is recommended that a Project Management Plan be developed during the early stages of every major project. The purpose of the plan is to clearly define roles, responsibilities, processes, and activities, which will result in the project being completed on time, within budget, with the highest degree of quality, and in a safe manner. âThe Scope of Work should be clearly defined in the Plan, along with change order and claims controls and other cost containment strategies to be used throughout the life of the project. Upper management buy-in from all sponsoring agencies should also be included in the form of a signature page.â Risk Charter A risk charter is similar to a list of red flag items. It is a list of identified risks that may be encountered during the life of the project. However, a risk charter is typically based on a more scientific assessment of risk, rather than simple engineering judgment. The charter may address the likelihood of the risk, the cost and schedule implications of the risk, and mitigation technique suggestions, as well as identifying which risks can have the largest impacts on the project. The goal of the risk charter is to reduce the number of risks on the list to as few as possible, by mitigation strategies or by project design changes. This method may be more effective than simply listing the potential problem areas, as with the red flagging of items, since it seeks to successfully eliminate the number of risk items.
46 Estimation Checklist Some DOTs use estimation checklists to ensure an estimate includes important items that frequently occur in projects. Checklist can help prevent the failure to include project items that might be needed, but are not yet designed at the time the estimate is completed. The level of detail in a checklist should mirror the detail of the estimate at any given level of project development. In the early phases of project development for example, checklists may be extremely simple; they then become more complex as the project advances through the development phases to correlate with more detailed definition of scope. One example checklist, used by a DOT during early project programming, includes the following: Functional/Preliminary Estimate List: 1. Clearing and Grubbing (acr. or ha.) 2. Earthwork (cy or m3) - unclassified, borrow, undercut, etc. 3. Fine Grading (sy or m2) 4. Drainage (per mile or kilometer) 5. Paving (ton or mtn, w/ pavement design, or sy/m2 without) 6. Stabilization (sy or m2) 7. Shoulder Drains (lf or meter) 8. Curb & Gutter (lf or meter) 9. Guardrail (lf or meter) 10. Anchor Units (each-type) 11. Fencing (mile or kilometer) 12. Interchange Signing (type and location) 13. Traffic Control (TCP) (per mile or kilometer) 14. Thermo and Markers (per mile or kilometer) 15. Utilities (lf or meters) 16. Erosion Control (acres or hectares) 17. Traffic Signals (each and location) 18. Retaining Walls / Noise Walls (sf or m2, with avg. height) 19. Bridges (individual location) 20. RC Box Culverts (individual location) 21. Railroad Crossing (each-with or without gates) Design Value Engineering Value engineering (VE) is used throughout the construction industry (SAVE International 2004). Within DOTs, VE is used to increase the project deliverables within the limited funds available for a project. By breaking the project into components, reviewing the function, and formulating
47 solutions and developing recommendations for improvements, one DOT has shown an increase in constructability, a minimization of ROW and/or environmental impacts, and a compression of construction schedules. Another DOT requires VE of all highway projects greater than $25 million. Design to Cost This is a technique that is used often in commercial development or manufacturing where a project must produce a product that will in essence pay for itself within a specified duration. A DOT can think of this as the matching of fuel tax revenue stream to the construction program; a processes used by some MPOs. In the private sector, a project must not only make a profit for the company but must also meet a minimum rate of return to be considered viable.7 The steps in this process require the company to set a target cost of the product, which includes the costs to make the product and the gross profit margin. Then the project team must evaluate the design alternatives and their costs. The design cost estimate and the target cost of the project are compared. If the estimated cost during design exceeds the target cost of the project, then one or both need to be re-evaluated before continuing with project development (Burman 1998). One DOT mentioned using an approach similar to this process. Rather than estimating the target cost, the DOT looks at the economic benefits generated by the highwayâs construction. In essence, the DOT cannot spend more on the road than the incoming revenue stream for its funding sources. Management of ROW, Utilities, and Environmental Issues The costs of various project items that are included in the estimate must be managed in different ways. One DOT specifically breaks out the various elements of an estimate in an effort to manage the costs of these elements more closely. The first tier of element definition is: PE Final design ROW Utility Construction A study phase as needed. Another DOT uses in-house software to manage the costs of various components. A checklist within the system requires each item to be entered by the project or task manager. In this system, ROW is completed by a ROW person in the district who can use their own numbers for the estimate or the model numbers based on the amount of ROW and current land use. The manager of the project must be careful that the level of definition for each element is consistent throughout the estimate. One DOT has a ROW division, which produces extremely straightforward and accurate estimates. This DOT has a training program for the ROW appraisers. 7 Interview with Coors Brewing Company, July 30, 2004.
48 State Estimation Department Numerous DOTs develop their project estimates at the central office. Locating estimation knowledge in one location provides the estimators with the ability to focus solely on estimation and maintenance of the cost databases that support estimate preparation. Through the use of one standardized process, DOTs can ensure estimate consistency. Cradle to Grave Estimators Maintaining the same estimator throughout all phases of project development allows that person to become intimate with the project scope and any unique characteristics impacting the project. This connectivity can increase estimate preparation efficiency as the estimator has historical knowledge about the project. In the case of one DOT that uses the same estimator from the start of project development through the final estimate, the projects are estimated at the local level. By keeping estimate preparation responsibility at the local level there is the advantage that the people doing the estimate understand the local situation and the political climate. Disadvantages of this system are that estimates are not always consistent throughout the state and estimation knowledge and costs data tends to be local. Year of Construction Costs Project cost estimates can be priced in current dollars or in year of construction dollars (cost inflated to the expected midpoint of construction date). The advantage of using year of construction cost is that it will more accurately reflect the cost of the project when it is complete and the communication of estimate is more credible. The disadvantages of using future costs are that it is difficult to compare current project costs with other projects in the STIP and inflated costs are not always accurate due to variances in inflation rates, market forces, and actual dates of construction. Scope Change Form Requiring completion of a scope change form for each change to the project permits tracking of scope changes as well the effects that the changes have on the project cost and schedule. Requiring scope change approval is extremely effective in managing project changes. The use of a form creates a discipline and awareness of the cost impact of scope changes. By notifying project team members in a timely manner of scope changes and their impact on project costs and schedule, appropriate actions can be taken to mitigate the impact of the change to the project and to other projects in the program. Several DOTs mentioned very rigorous scope change systems requiring certain forms be completed and approvals from or notification of specified personnel, if a scope change is made. Cost Containment Table The Cost Containment Form shown in Table 3.2 requires updating at each predetermined project milestone. At each project milestone point where this form is used the estimate must be broken down by specified items. The form has space available for scope definition as well as comments. A change in an estimate category is evident when completing the form and this allows for immediate investigation and notation of explanations. This effort to manage project costs continues from the programming and advanced planning/preliminary design stage through final design until the project letting.
49 Table 3.2. Cost Containment Form used by One DOT Cost Containment Table District: Program Yr: County: Project: Short Title: Cost Containment Milestone Estimate Cost Breakdown Program Amount (PMC approved amount) $ E&E Scoping Field View $ 30% (Design Field View) $ 75% (After Final Design Field View) $ 95% (Engineer's Estimate) $ Bid Amount $ Engineering: Preliminary Engineering Final Design R/W Utilities Construction Total Cost: Scope Comments Gated Process In essence, a cost containment spreadsheet, such as the Form shown in Table 3.2 which was discussed previously, creates a gated development process because projects cannot move from one milestone to the next without approval of the Cost Containment Form. One non- transportation source communicated that they use a gated process, which is extremely formalized. Before a project can continue in the development process the project team must hold a meeting in which the Construction Industry Instituteâs Project Definition Rating Index (PDRI 1997) must be completed. The PDRI scores a projectâs level of scope definition as compared to a historic data on scope definition. The project must achieve a maximum score before the project can continue. If the project does not obtain the maximum score then the project is returned to the previous phase for more definition. The PDRI requires that the project be scored in of the following areas; Basis of Project Decision a. Manufacturing Objective b. Business Objective c. Basic Data Research and Development
50 d. Project Scope e. Value Engineering Front End Definition Site Information Process/Mechanical Equipment Scope Civil, Structural, and Architectural Infrastructure Instrument and Electrical Execution Approach Procurement Strategy Deliverables Project Control Project Execution Plan Create Project Baseline All DOTs need to establish a baseline scope and estimate for their projects. The project baseline scope and estimate is used to measure performance throughout project development and construction. This baseline may be created at different points by different DOTs but the purpose is the same; it defines the moment when an identified need becomes a ârealâ project and is budgeted. Some DOTs establish a baseline estimate early in the project development process, during long range planning, where other DOTs set the base line when the project is programmed or at 35% design. Estimation Manual The creation of a DOT specific estimation manual helps to ensure consistency in estimate preparation. Some DOTs have estimation manuals that pertain to estimates starting at the earliest phases of project development, while other DOT manuals do not indicate how estimates should be prepared until the later phases of project development. A few DOTs do not have any type of formal estimation manual. The estimation manuals that are available vary considerably in depth and quality of provided information. The manuals for several DOTs are very general while others provide great detail regarding preparation and content of the estimate. Estimator Training Very few states offer formal training to their estimators. Many states noted that the training that takes place must either be requested or occurs on the job. The formalized training that is present
51 by DOTs does not necessarily address all portions of the estimate preparation. One DOT mentioned that there is formal training for ROW estimators, while there is no formal training for the personnel that estimate the remainder of the project. Other DOTs have training courses that teach the estimators how to use the software and expose the participants to details and issues that they should consider when creating cost estimates. Estimation Scorecard Estimation scorecards can be used to measure the success of project development processes. These scorecards are developed early in the project development process. Scorecards are most commonly used when consultants are preparing the project design and estimate. They indicate the measures that will be used at project completion to evaluate success. Once the project is completed the consultants fees can be based off of target values designated during project development and the achieved values measured after project completion. A set of scorecards is developed for each project, one for execution and another for benefit. The elements of the execution scorecard for determining project success are cost, schedule, and quality/performance. The benefit scorecard elements are defined based on the project. Each scorecard element is measured as either above target, on target, or below target. Early identification of the project success measures ensures that there is no miscommunication regarding functionality and physical structure of the completed project. This helps to clearly align defined project scope with expectations thereby limiting scope changes. Final Design Estimation and Management Software Similar or the same software used during the earlier project development phases is used for preparing the final estimate, and again it can also be used to manage project cost. Commercially available software is capable of this function as well as DOT developed software. Currently many DOTs use the Trnsâ¢port modules at this level, even if in-house programs were used in the previous phases. Estimate Review The review of project estimate at this stage can vary from none, to an in-house/peer review, to a formal committee review. The less formal review can include another estimator in the state estimation office or design division, who examines the estimate before the project is bid. This review may only check to make sure that no items were missed. This review is typically based on experience or a formal check system. In many cases during this phase, DOTs have more formal estimate reviews, which require the estimate be presented to a committee. The committee can consist of a number of people including department heads and field personnel representing the state construction engineer, Federal Highway Administration, the contract administration engineer, the state maintenance engineer, and/or the project/field engineer. The committee may ask for more information regarding elements of the estimate. The committee then votes regarding approval of the estimate. SUMMARY OF IMPORTANT ISSUES Cost estimation practices and management techniques describe by the DOTs to the research team attempt to alleviate many causes of cost escalation. However, it appears that no single DOT has
52 cost estimation practice and cost estimation management systems in place that address all factors causing cost escalation that Departments must address in establishing and managing the cost of their projects. Contingency and Uncertainty Contingency is typically applied to DOT cost estimates but its application must still be considered a deficiency. It was found that in most DOTs the application of a contingency to an estimate is so loosely defined that typically there is no consistent application of contingency. The DOTs are aware that potential issues exist for each project and therefore incorporate contingency. However, they very often fail to define the specific aspects contingency dollars are supposed to cover. To a large extent the problem is the result of the fact that contingency means what the estimator says that it means. As a result, issues that should not be a part of contingency consume the contingency budget leaving no funds for its intended purpose. By definition contingency is meant to cover: 1) an event that may occur but that is not likely or intended or 2) a possibility that must be prepared against, the condition being dependent on chance. Often the amount of contingency added to an estimate is dependent on engineering judgment rather than an analytical approach causing inconsistent application of contingency. Risk-Based Estimation and Management Risk-based estimation and management is used by only a small number of transportation agencies. Range estimates and risk charters are common practice in the other industries, but the highway sector is just beginning to apply these techniques. The DOTs who are applying a risk- based estimation approach have found it to be successful in communicating the true nature of project costs at the planning and preliminary design phase. These DOTs have also found it useful in managing the project development and design process. Time Value of Money Many DOTs inflate their estimates to the prospective date of construction by applying a factor that reflects the current economic situation. However, DOTs do not usually consider the impact of a schedule change on inflation. Prolonging the schedule will increase the cost of construction. For example, a million dollar project that has been postponed for one year would experience an additional $30,000 in cost if the current inflation factor were 3 percent. If the estimates are periodically reviewed the schedule must also be considered, then the DOT might consider the impact of time changes and incorporated it into the estimate. However, many departments do not have a regular formal estimate review process. Scope Control The issue of scope control is paramount to managing project costs. DOTs are attempting to use a variety of methods and tools to control scope changes8 and scope growth, but their use is not widespread and it is far from standard practice. Scoping documents are being used in the project development process to identify and specify critical design elements. Definitive management 8 By Arizona law (A.R. S. 28-6353) if there is a material scope change for freeways in Phoenix the change must be approved by the Maricopa Association of Governments.
53 plans are being recommended on major projects. Cost containment tables are being used to identify when a projectâs scope has grown, but these tables may not indicate growth until significant growth has occurred. Project Baseline and Gated Process A system of cost validation points must be established if a project is to remain on budget. Few DOTs have a process, which is gated based upon estimated cost. A gated process begins with a standard point in design in which a project estimate baseline is created. Some DOTs establish a baseline estimate early in the project development process, during long range planning, where other DOTs set the base line when the project is programmed or at 35% design. However, many do not have a standard point for setting a project baseline at all. Upon establishment of the project estimate baseline, a gated process involves the validation of scope, schedule and cost at critical points during the project development process. A project should not be allowed to move to the next phase of development unless all scope, schedule and cost constraints have been established. Estimate Reviews Most of the DOTs have informal reviews that are conducted by the project team. Frequently the individual preparing the estimate is responsible for the quality of the estimate. As a result, the DOTs rely on the individualâs judgment to impartially review the estimate. Although the final project estimate is reviewed before letting, periodic reviews and approval are seldom required during the projectâs development. Reviews typically occur after the projectâs cost has increased or a major scope change has occurred. A few DOTs have requirements that an estimate (the projectâs estimated cost) must remain within an established range. If the estimated cost goes outside the range, then additional reviews and approvals are needed. The informality of the review process leads to projects advancing to the next stage without serious cost reviews. Estimator Qualifications The reliance on estimators who lack sufficient experience is another deficiency that DOTs must overcome. When an experienced estimator retires or moves to another job that estimatorâs knowledge is lost. This dependency hinders the DOTsâ because they rarely have a training program for new estimators or an estimation procedure documented (a Manual) with sufficient detail for an inexperienced individual to follow. Estimation Documentation Proper estimation documentation is another common deficiency, which causes accountability issues. Unless a DOT has to request additional funding, the reasons that cause a project cost increase or a scope change is not recorded and therefore not traceable. Many DOTs lack consistent estimation procedures between their districts. Many DOTs do not have standardized estimation procedures, and they allow the districts to use whatever approach the Districts deem suitable. Management cannot properly correct a problem if they do not know how an estimate was prepared or what changes were made during project development.
54 Communication The DOTs also lack coordination and communication between the disciplines participating in the development of the projectâs scope and estimate. The supporting groups, who feed information into the primary estimate, do not play an active role in the projectâs development until the project reaches the preliminary design phase. In addition, the individual who is compiling the estimate is often not certain the other groups have properly accounted for their project elements. Project Complexity Most of the DOTs do not adequately consider project complexity when they create a cost estimate. If a project is more complex than the DOTâs standard projects, then the DOT might include additional contingency. During a project review, some DOTs consider complexity by requiring more approval signatures than a less complex project, but the impact a highly complex project has on the cost estimate is not considered. CONCLUSIONS This chapter critically reviewed current practices in the area of cost estimation practice and cost estimation management. A number of unique or innovative approaches to cost estimation practice and cost estimation management were described. A discussion of how current cost estimation practice and cost estimation management approaches do and do not address the identified potential root causes of project cost escalation was also provided. Finally, important issues and more specifically deficiencies in current practice are identified. Strategies to address these and other issues are presented in Chapter 4, including proposed methods and tools to implement the strategies.
