DSpace at VNU: Case Law and Variations in Cumulative Impact Productivity Claims

Case Law and Variations in Cumulative Impact Productivity Claims Downloaded from ascelibrary.org by New York University on 05/10/15 Copyright ASCE For personal use only; all rights reserved Long D Nguyen1 and William Ibbs, M.ASCE2 Abstract: Proving and quantifying lost productivity due to cumulative impacts of multiple changes are difficult tasks This paper presents the most acceptable methods from case law and demonstrates their applications for analyzing the loss of productivity These methods include earned value analysis, measured mile analysis, and combinations of these two They are either well established or drawn from recent court and board decisions A case study is used to illustrate and compare the use of these methods These methods result in considerably different loss of productivity values though the actual amount ͑i.e., inefficiency in labor hours͒ is unique for a particular case and though these methods are often thought to be similar or even the same How a measured mile analysis and its variants are employed affects the amount of lost productivity estimated The variants can avoid some drawbacks of measured mile and earned value studies Nevertheless, which method is more accurate and reliable is difficult to provide for a particular claim Practitioners should choose between them based on the availability of project records and the nature of changes and cumulative impacts Practitioners may also employ two or more methods to perform a “sensitivity analysis” of the chosen methods and persuade the other party and/or the jury that their estimate of lost productivity is sufficiently certain DOI: 10.1061/͑ASCE͒CO.1943-7862.0000193 CE Database subject headings: Claims; Contracts; Court decisions; Productivity; Owners; Contractors Author keywords: Claims; Contracts; Court decisions; Productivity; Owners; Contractors Introduction Change can be disruptive to construction Multiple change orders may cause a cumulative impact or ripple effect on labor productivity of unchanged work Previous studies have shown the negative impact of changes on labor productivity ͑Leonard 1988; Ibbs and Allen 1995; Hanna 2001͒ The U.S courts and boards of contract appeals have acknowledged cumulative impacts claims for a few decades In Centex Bateson Construction Co ͑1998͒, cumulative impact is described as “the unforeseeable disruption of productivity resulting from the ‘synergistic’ effect of an undifferentiated group of changes.” Liability, causation, and resultant injury are the three prerequisite components for a successful cumulative impact claims Liability consists of a legal right to recover pursuant to the contract or due to the breach of the contract by owner and evidence of owner-caused disruptions ͑i.e., a multitude of change orders͒ ͑Jones 2003͒ As to causation, the contractor must prove the inefficiency caused by the owner’s changes Finally, the contractor Lecturer, Faculty of Civil Engineering, Ho Chi Minh City Univ of Technology, Vietnam; formerly, Construction Consultant, Jax Kneppers Associates, Inc ͑JKA͒, Walnut Creek, CA 94598 E-mail: ndlong@ hcmut.edu.vn Professor of Construction Management, Dept of Civil and Environmental Engineering, Univ of California, Berkeley, CA 94720; and, President, The Ibbs Consulting Group, Inc., Oakland, CA 94618 E-mail: William.Ibbs@ibbsconsulting.com Note This manuscript was submitted on May 7, 2009; approved on January 25, 2010; published online on January 28, 2010 Discussion period open until January 1, 2011; separate discussions must be submitted for individual papers This paper is part of the Journal of Construction Engineering and Management, Vol 136, No 8, August 1, 2010 ©ASCE, ISSN 0733-9364/2010/8-826–833/$25.00 needs to present a reasonable estimate of the loss of productivity caused by changes to fulfill the last component While the liability can be straightforward, causation and resultant injury may be more difficult to prove This paper presents methods for proving and quantifying the loss of productivity based upon the project data While a few methods are well established, others that are drawn from recent court and board decisions may be more obscure Nevertheless, it is instructive to examine all these methods and such is done so in this paper through a case study The focus of this paper is such project-specific studies as earned value analysis, measured mile analysis, and combinations of the two can result in significantly different estimates of loss of productivity Project Specific Studies and Productivity Analysis Many methods for estimating lost labor productivity are available They are grouped into project-specific studies, project comparison studies, specialty industry studies, general industry studies, cost basis, and productivity impact on schedule ͓Association for the Advancement of Cost Engineering ͑AACE͒ International, Inc 2004͔ Among them, approaches in project-specific studies are most desirable In addition, measured mile and earned value analyzes in this group are the most accepted by courts and boards Improved measured mile studies have recently been proposed, for example, baseline productivity analysis ͑Thomas and Završki 1999; Thomas and Sanvido 2000; Gulezian and Samelian 2003͒ and statistical clustering ͑Ibbs and Liu 2005͒ However, they seem not to have been tested by a court or board Discussion of these improvements is out of the scope of this paper Measured mile analysis determines lost productivity by comparing the unit productivity costs in an undisrupted time period or 826 / JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / AUGUST 2010 J Constr Eng Manage 2010.136:826-833 Table Cases Denying Cumulative Impact Claims Downloaded from ascelibrary.org by New York University on 05/10/15 Copyright ASCE For personal use only; all rights reserved Case Reason for rejection Remark Watt Plumbing, Air Conditioning and Elec., Inc v Tulsa Rig, Reel and Mfg Co ͑1975͒ Not breach of contract because modification of contract becomes part of contract and modified contract was agreed by the parties Contractor alleged breach of contract by excessive change Dyson and Co ͑1978͒ Contract language not allowing for claims not included in the previous change orders 19% increase in contract value due to 39 change orders Pittman Constr Co v United States ͑1983͒ Not a fundamental change in the contract 12% increase in contract value due to 206 change orders; 10% increase in duration Central Mechanical Constr ͑1986͒ Contractor failed to reserve the rights to recover cumulative impact costs Release language included in change orders Freeman-Darling, Inc ͑1989͒ Not a fundamental change in the contract for a cumulative impact Numerous change orders Triax Co v United States ͑1993͒ No proof of causal linkage between numerous changes and cost overruns The court agreed the impact of changes on labor productivity The Appeal of Gulf Coast Trading Co ͑1994͒ No proof of causation shown Differing site condition and reduced production rate Southwest Marine, Inc ͑1994͒ Not a fundamental change in the contract and insufficient causal connection 202 change orders were issued during the course of work Dawson Constr Co ͑1993͒ No proof of causation Numerous change orders Centex Bateson Constr Co., Inc ͑1998͒ No proof of causation The Board agreed that cumulative impacts may be recovered but held that the existence of multiple change orders is not sufficient The Appeal of Coates Industrial Piping, Inc ͑1999͒ No proof of causal linkage between changes and loss of labor efficiency Permit delay; and design deficiencies; 157 CORs Fru-Con Constr Corp v United States ͑1999͒ The burden of proving the causation and amount of productivity loss with sufficient certainty failed Lost productivity claims due to differing site conditions and abnormal adverse weather physical area to that spent in the alleged disrupted time period or physical areas ͑Zink 1986͒ The measured mile is a continuous time period in which the productivity is not affected by factors caused by the owner ͑Thomas and Sanvido 2000͒ An advantage of this method is the use of the actual performance rather than the original estimate for quantifying lost productivity For that reason, the measured mile is quite credible Earned valued analysis can also be employed for estimating inefficiency It is especially effective when productivity data are not available for a measured mile analysis This is because insufficient information, such as a lack of physical units of work installed, can make productivity measurement difficult ͑AACE International, Inc 2004͒ This technique is based on a realistic budget for the contract, which is not always the case and relies on a method for measuring percent complete being acceptable to the disputing parties Combinations of earned value and measured mile analyzes have been proposed, applied and recognized by courts and boards This is to reduce the shortcomings of these two individual methods The measured mile analysis technique that requires identical or substantially similar work for productivity comparisons can limit its applicability for unique and complex tasks ͑Loulakis and Santiago 1999͒ This technique is also inappropriate when a measured mile period or actual productivity records are unavailable In contrast, an unreasonable estimate diminishes the reliability of the earned value analysis technique Therefore, several variants, for example, the combinations of earned value and measured mile analyzes, have successfully been used in cumulative impact claims Cases with Cumulative Impact Claims Courts and boards of contract appeals have recognized cumulative impact claims since late 1960s In Bell BCI Co v United States ͑2008͒, the U.S Court of Federal Claims confirmed that: “Until 1967, the Rice doctrine ͑Rice v United States 1942͒ precluded courts from considering the effect of change orders on portions of the work not directly covered by the change … To avoid inequitable results, the Government changed the standard Changes clause in late 1967 to add language that covers the effect of changes on unchanged work.” Specifically, cumulative impact costs may be awarded if the three components of liability, causation, and resultant injury are proved with sufficient certainty as previously discussed A review of cases is needed to identify reasons why cumulative impact claims are successful or unsuccessful in legal proceedings Tables and present representative cases that courts or boards denied and allowed cumulative impact claims, respectively The cases in each table are listed chronologically In Table cumulative impact claims were denied due to any of the three typical reasons: release/waiver language in either the contract or contract modifications ͑cases #1, 2, and 4͒; changes not resulting in a fundamental change in the contract ͑cases #3, 5, and 8͒; and no or insufficient proof of causal connection between lost productivity and changes directly or constructively ordered by the other party ͑i.e., owner͒ ͑cases #6, 7, 8, 9, 10, 11, and 12͒ In Central Mechanical Construction ͑1986͒, the Board rejected a cumulative impact claim because release language in change orders aimed at covering all direct and indirect costs and therefore barred any subsequent claim In Pittman Construction Co ͑1983͒, a cumula- JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / AUGUST 2010 / 827 J Constr Eng Manage 2010.136:826-833 Table Cases Allowing Cumulative Impact Claims Downloaded from ascelibrary.org by New York University on 05/10/15 Copyright ASCE For personal use only; all rights reserved Case Reason for allowance Remark Electronics and Missile Facilities v United States ͑1969͒ Lost productivity caused by changes on earthwork Initially denied by a BCA Litton Systems, Inc., Ingalls Shipbuilding Division ͑1978͒ A fundamental change in the contract caused by multiple change orders 58% increase in contract value due to change orders Coley Properties Corp v United States ͑1979͒ Unchanged work was performed out of sequence due to numerous change orders The impact of changed work on unchanged work C Norman Peterson Co v Container Corp ͑1985͒ Breach of contract and lost productivity due to multiple errors and design changes Hundreds of change orders but not in writing State ex rel Dept of Transp v Guy F Atkinson Co ͑1986͒ Ongoing piecemeal changes impacted the entire operation Jury verdict method was adopted because sufficient proof of damages was impossible David J Tiernay ͑1988͒ Lost productivity due to multiple change orders Jury verdict method was adopted due to insufficient proof of damages Charles G Williams Constr., Inc ͑1989͒ Compensation for cumulative impact costs 26 change orders were issued due to defective drawings and specifications Atlas Constr Co., Inc ͑1990͒ Recognized cumulative impacts Productivity loss as a result of multiple change orders Additional costs for updating CPM schedules and field office overhead were also awarded Clark Concrete Contractors, Inc v General Services Administration ͑1999͒ Damages for increased labor costs due to lost productivity, additional work, and overtime premium Redesign was ongoing during the course of construction The Clark Construction Group, Inc ͑2000͒ Undeniable productivity losses caused by change in construction sequence and wet conditions Construction sequence changed from horizontal to vertical; late response to RFIs; design conflicts The Appeal of P J Dick, Inc ͑2001͒ Electrical design deficiencies and constructive acceleration reduced electrical ͑branch circuit͒ labor productivity Measured mile of similar work ͑feeder circuit͒ was allowed Amelco Electric v City of Thousand Oaks ͑2002͒ Breach of contract Noticeably, the sheer number of changes did not result in abandonment of a public works contract Contractor failed to distinguish inefficiencies caused by different parties ͑contractor, owner/construction manager, etc.͒ Bell BCI Co v United States ͑2008͒ Reasonable basis for a cumulative impact claim 34% increase in contract value due to 206 contract modifications; 730 extra work orders; 70% increase in duration tive impact claim was denied because the Board held that the total of changes ͑12% increase in contract value͒ was not a fundamental change in the contract The U.S Court of Federal Claims refused a lost productivity claim in Triax Co v United States ͑1993͒ and reasoned that the claimant failed to prove the causal connection between numerous change orders and lost productivity The total cost method was used for calculating cumulative impact costs in this case Courts and boards typically awarded cumulative impact claims with sufficient demonstration that an excessive number of changes occurred and caused a ripple impact on efficiency ͑Table 2͒ Although measured mile analysis is popular, other methods have been used successfully in cumulative impact cases In C Norman Peterson Co v Container Corp ͑1985͒, quantum meruit was used to calculate the reasonable value of contractor’s work performed because of numerous change orders and contract abandonment Quantum meruit is “the measure of damages for recovery on a contract that is said to be ‘implied in fact’” ͑McConnell 1997͒ In The Clark Construction Group, Inc ͑2000͒, the Board awarded cumulative impact costs to the contractor and its subcontractors in the nature of a jury verdict by using the productivity factors from the Mechanical Contractors Association of America ͑MCAA͒ Manual to estimate productivity losses Preferred methods with project-specific cost and/or productivity data were used successfully in many of the above cases ͑Table 2͒ In Bell BCI Co v United States ͑2008͒, a combination of earned value and measured mile analyzes ͑i.e., earned hours ver- sus unearned hours͒ showed 80,317 h of the 320,703 total hours lost due to the cumulative impact of change orders and satisfied the court: “Through 1999, Bell had completed a significant portion of the base Contract work … and incurred a nonproductive work rate of eight percent on the project … Mr Brannon attributed one-half of the nonproductive hours to Bell for this period, and set Bell’s reasonable productivity level at 104 percent of the originally planned productivity … Mr Brannon compared actual hours incurred on the project against 104 percent of the originally planned hours to calculate unearned hours attributable to the NIH changes Mr Brannon found that approximately 25 percent of Bell’s total hours expended on the project were attributable to labor productivity loss caused by the NIH changes.” Another combination was observed in James Corp d/b/a James Construction v North Allegheny School District, et al ͑2007͒ However, the parties and court in this case seemed to name that analysis as measured mile: “Dividing the Project into two time periods, claim expert compared the percentage of work completed in each period to the number of labor hours utilized During the first period, Contractor expended 4,279 hours to complete 41.76% of the Project Had Contractor been able to work at the same pace during the second period, it would have expended an additional 5,967 hours to complete the remainder of the Project However, the Project’s total hours equaled 19,645 hours; therefore, Contractor used 15,366 hours to complete the second period ͑19,645—5,967͒ Thus, the 828 / JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / AUGUST 2010 J Constr Eng Manage 2010.136:826-833 Table Summary of Contractor’s Records Planned Month Downloaded from ascelibrary.org by New York University on 05/10/15 Copyright ASCE For personal use only; all rights reserved ͑1͒ Actual Earned Quantity Productivitya Labor hours Labor costs Quantity total Quantity COs Productivity Labor hours Labor costs Labor hours Labor costs ͑2͒ ͑3͒ ͑4͒ ͑5͒ ͑6͒ ͑7͒ ͑8͒ ͑9͒ ͑10͒ ͑11͒ ͑12͒ $67,778 $110,909 $185,440 $258,077 $254,167 $388,182 $610,000 $449,474 $273,684 $61,905 $56,522 $28,261 $2,744,398 909 1,739 3,167 4,583 4,167 5,833 6,667 5,833 3,333 833 909 455 38,429 $55,455 $106,087 $193,167 $279,583 $254,167 $355,833 $406,667 $355,833 $203,333 $50,833 $55,455 $27,727 $2,344,140 2/08 1,000 1.1 909 $55,455 1,000 0.9 1,111 3/08 2,000 1.15 1,739 $106,087 2,000 1.1 1,818 4/08 4,000 1.2 3,333 $203,333 3,800 1.25 3,040 5/08 5,000 1.2 4,167 $254,167 5,500 1.3 4,231 6/08 5,000 1.2 4,167 $254,167 5,000 1.2 4,167 7/08 5,000 1.2 4,167 $254,167 7,000 3,500 1.1 6,364 8/08 5,000 1.2 4,167 $254,167 8,000 3,500 0.8 10,000 9/08 5,000 1.2 4,167 $254,167 7,000 1,500 0.95 7,368 10/08 3,000 1.2 2,500 $152,500 4,000 800 0.95 4,211 11/08 1,000 1.1 909 $55,455 1,000 1.05 952 12/08 500 1.1 455 $27,727 1,000 1.15 870 1/09 500 1.15 435 Total 36,500 30,679 $1,871,390 45,800 9,300 44,566 a Labor rate was $61 ͑$65͒ per hour for period of February to October 2008 ͑November 2008 to January 2009͒ inefficient labor hours amounted to 9,366 ͑15, 366– , 967 = , 366͒ To arrive at an earned value factor of 61%, claim expert divided the inefficient labor hours by the number of hours worked in the second period ͑9 , 366÷ 15, 366= 61%͒ In sum, because the existence of damages was established, because the measured mile analysis offers a reasonable basis upon which damages can be calculated, and because the trial court found this approach persuasive, we discern no merit in School District’s assertions.” Courts and boards allowed measured mile of both same work and similar work In Clark Concrete Contractors, Inc v General Services Administration ͑1999͒, the board upheld the contractor’s measured mile of the same activities of concrete work The measured mile established the production rates in unimpacted period ͑before redesign process͒ and in impacted period for four activities: forming of the floor slabs; stripping of slab formwork; combined forming and stripping of columns; and combined forming and stripping of stairs: “To determine the labor inefficiency costs resulting from the blast changes in the forming of the concrete floor slabs, OMNI identifies as an unimpacted period the time from the beginning of formwork until August 28, 1995, the date on which the contractor received the first of the massive blast design changes The time from August 28, 1995, to March 10, 1996, when the forming of the third floor slab was completed atop the added blast wall, is a highly impacted period The time from March 10 to July 7, 1996, when slab forming activities ended, is a lesser impacted period… OMNI achieved the following productivity rates: during the unimpacted period, 048 man-hours per square foot; during the highly impacted period, 113 man-hours per square foot; and during the lesser impacted period, 065 man-hours per square foot.” In The Appeal of P J Dick, Inc ͑2001͒, productivity of branch circuit work was allegedly lost due to design deficiencies Because no unimpacted period was identifiable for branch circuit work, the claimant used installation of feeder circuit as similar work to support its measured mile analysis The measured mile analysis using the feeder-branch circuit comparison resulted in, of the 70,498 actual branch circuit labor hours, 39,354 unproductive labor hours attributable to owner-caused inefficiency: “He ͓contractor’s expert͔ divided the actual man-hours worked on feeders by the adjusted budget for feeders to arrive at a “dem- onstrated efficiency” factor of 1.147 That is: To earn each manhour in the budget for the feeders, Kent Electric had to work 1.147 h … Multiplying this budget by the demonstrated efficiency factor of 1.147, results in an adjusted branch budget of 31,143.8 man-hours Deducting this adjusted budget from the actual branch hours of 70,498, results in 39,354.2 unproductive man-hours At the stipulated hourly wage rate, the unproductive man-hours results in additional labor costs of $1,471,453 attributable to VA ͓owner͔ caused inefficiency.” Hypothetical Case Study The above cases showed that Courts and Boards have allowed various approaches for analyzing causation and quantum of cumulative impact claims These approaches are adopted for productivity analyses in the following hypothetical case A contractor alleged lost productivity caused by cumulative impacts of multiple changes Construction work was originally scheduled for 11 months to complete 36,500 units ͑weight of structural steel͒ from February to December 2008 With labor rate of $61 per hour, total labor hours and labor costs were estimated 30,679 and $1,871,390, respectively During the period of July to October 2008, the owner directed many changes totaling to 9,300 units or $567,300 of labor costs, approximately 30% of the original contract work The approved time extension was one month The contract was completed on time ͑January 2009͒ Actual labor costs were $2,744,398 Therefore, the contractor’s labor budget was exceeded $305,708 ͑$2,744,398—$1,871,390—$567,300͒ Table compiles the contract data Assume that the contractor is entitled for cumulative impacts; the inefficiency costs need to be quantified with sufficient certainty Alternative Analyses of Lost Productivity Earned value and measured mile and their combinations are used to present lost productivity JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / AUGUST 2010 / 829 J Constr Eng Manage 2010.136:826-833 1,5 12.000 10.000 1,3 Units/Labor-hour Labor-hours 8.000 6.000 4.000 1,1 0,9 0,7 2.000 Downloaded from ascelibrary.org by New York University on 05/10/15 Copyright ASCE For personal use only; all rights reserved Planned labor-hours Actual labor-hours Earned hours Fig presents planned, actual, and earned labor hours ͑LHs͒ from columns 4, 9, and 11 ͑Table 3͒, respectively Earned value analysis can be used for quantifying the loss of productivity The difference between the actual hours expended and the earned hours for the period of the impact may be used to calculate the inefficiency experienced ͑AACE International, Inc 2004͒ That is, loss of efficiency can be calculated by n LHsk − Earned LHsk͒ ͑in LHs͒, or n Loss of efficiency = Jan-09 Dec-08 Nov-08 Oct-08 Sep-08 Aug-08 Jun-08 May-08 Jul-08 Actual Productivity Fig Planned versus actual productivity Earned Value Analysis ͑Actual ͚ k=1 Apr-08 Planned Productivity Fig Planned, actual, and earned hours Loss of efficiency = Mar-08 Feb-08 0,5 ͑Actual Costk − Earned Valuek͒ ͑in ͚ k=1 $͒ where k = kth month ͑day, week͒ of the impacted period and n = number of months ͑day, week͒ of the impacted period The actual labor hours are much greater than the earned hours during the four-month period of July to October 2008 while they are less different in the other periods ͑Fig 1͒ That four-month period coincided with the time multiple change orders occurred This indicates that multiple change orders caused lost productivity during that period with an assumption that other disruptions are negligible The actual and earned hours for this impacted period are 27,943 and 21,667 labor hours, respectively The inefficiency hours caused by cumulative impacts of change orders are therefore 6,276 labor hours Loss of efficiency determined by earned value analysis may be questionable An earned value analysis is doubtful unless the budget is substantiated ͑Ibbs et al 2007͒ Other project records have to be corroborated to confirm that the original estimate ͑i.e., planned productivity͒ and inefficient labor hours are reasonable For that reason, the following measured mile analysis and its variants are more reliable and more acceptable Measured Mile Analysis Measured mile analysis compares the productivity in impacted period and productivity in unimpacted period ͑Fig 2͒ The unimpacted period is March to June of 2008 As a “build-up” time, February 2008 is not included because it is not representative of expected productivity ͑Zink 1986͒ In Clark Concrete Contractors, Inc v General Services Administration ͑1999͒, the contractor successfully employed this method to prove the productivity loss of construction work caused by design changes: Fig demonstrates the planned and actual productivities over time The actual productivities in unimpacted periods were generally higher than those in the impacted period This proves that changes caused labor inefficiency Table shows the productivity analysis for the impacted period The expected productivity ͑1.23 units/labor hour͒ is the weighted average of productivities achieved in March to June of 2008 The labor inefficiency totaled to 6,858 labor hours Noticeably, the formula for the expected productivity and lost productivity in the impacted period is n Qk Actual ͚ k=1 ‫ء‬ Expected productivity = Productivityk n Qk ͚ k=1 n = Qk ͚ k=1 ͑in units/LH͒ n Actual ͚ k=1 LHsk Lost productivityi = Expected productivityi – Actual productivityi ͑in units/LH͒ Table Productivity Loss with Measured Mile Impacted period July 2008 August 2008 September 2008 October 2008 Total Actual productivity Expected productivity Lost productivity Quantity installed Inefficient labor hours 1.10 1.23 0.13 7,000 687 0.80 1.23 0.43 8,000 3,512 0.95 1.23 0.28 7,000 1,692 0.95 1.23 0.28 4,000 967 26,000 6,858 830 / JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / AUGUST 2010 J Constr Eng Manage 2010.136:826-833 Downloaded from ascelibrary.org by New York University on 05/10/15 Copyright ASCE For personal use only; all rights reserved Table Inefficient Labor Hours with Method Used in James Corporation d/b/a James Construction v North Allegheny School District, et al ͑2007͒ Description Quantity % complete before the impact ͑February to June 08͒ Actual labor hours before the impact % complete during the impact ͑July to October 08͒ Expected labor hours during the impact Actual labor hours during the impact Inefficient labor hours due to the impact 37.77% 14,367 56.77% 21,592 27,943 6,351 Loss of efficiencyi = Q‫ء‬i ͩ 1 − Actual Productivityi Expected Productivityi ͪ ͑in LHs͒ where k = kth month ͑day, week͒ of the measured mile ͑unimpacted͒ period; n = number of months ͑day, week͒ of the measured mile period; i = ith month of the impacted period; and Qi = actual quantity installed at month i Compared to earned value analysis, the measured mile study is more reliable in this instance The expected productivity is calculated based on the actual productivities achieved, not estimated as in earned value analysis In other words, the above earned value analysis implicitly compares planned productivity and actual productivity in the same impacted periods to find the loss of productivity Earned Value and Measured Mile Combined Courts and Boards of Contract Appeals have accepted analyzes that combine earned value and measured mile Though these courts and boards still termed it “measured mile” for these analyzes, by definition, they are not a traditional measured mile analysis as defined in the previous section, but rather variants That is, they are not a comparative productivity analysis used to establish lost productivity The following examples compute inefficient labor hours for the case study using methods accepted in recent court rulings From James Corporation d/b/a James Construction v North Allegheny School District, et al „2007… This variant compares the percentage of work performed to the number of labor hours used The project is divided into three periods, before the impact ͑February to June 2008͒, during the impact ͑July to October 2008͒, and after the impact ͑November 2008 to January 2009͒ The contract spent 14,367 labor hours to complete 37.77% of the project by June 2008 ͑Table 5͒ The percent complete for a particular period equals to the quantity installed in that period divided by the total quantity installed Similarly, the contractor spent 27,943 labor hours to complete 56.77% of the project during the impacted period Had the disruption not occurred, the contractor would have spent only 21,592 labor hours ͑14, 367‫ء‬56.77% / 37.77%͒ to complete this portion of the work The inefficient labor hours caused by the multiple changes are 6,351 ͑27,943—21,592͒ The mechanics for calculating the loss of productivity is below Expected LHs in impacted period = Actual LHs in unimpacted period ‫ء‬ ͩ ͪ b% ͑in LHs͒ a% Loss of efficiency = ͑Actual LHs – Expected LHs͒ in impacted period ͑in LHs͒ where a and b = percent complete of the unimpacted and impacted period, respectively Noticeably, to strictly apply the analysis used in the case of James Corporation d/b/a James Construction v North Allegheny School District, et al ͑2007͒, the project can also be divided into two periods, before the impact ͑February to June 2008͒ and during and after the impact ͑July 2008 to January 2009͒ Had the disruption had not occurred, the contractor would have spent additional 23,668 labor hours ͓14, 367‫͑ء‬100% − 37.77%͒ / 37.77%͔ to complete the project The actual labor hours for the remaining work were 30,199 A similar analysis yields $6,532 of inefficient labor hours From Bell BCI Co v United States „2008… This variant starts with identifying “reasonable labor-hour level” as a ratio of the actual and planned labor hours for the planned quantity installed in the unimpacted period ͑February to June 2008͒ It then identifies reasonable labor hours for the impacted period and compares them with the actual labor hours Though the case of Bell BCI Co v United States ͑2008͒ named “reasonable productivity level,” “reasonable labor-hour level” is used to avoid confusion That is, in Bell BCI Co v United States ͑2008͒, “reasonable productivity level” was determined at “104 percent of the originally planned productivity” in the unimpacted period One may think that the actual productivity was higher than the planned but in fact the actual productivity was less than the planned productivity in the unimpacted period The following is the process for quantifying the productivity loss: Reasonable LH level = ͩ ͪ Actual LHs in unimpacted period ‫ ء‬100% ͑in%͒ Budgeted LHs in unimpacted period Reasonable LHs in impacted period = ͑Earned LHs in impacted period͒‫͑ء‬Reasonable LH level͒ Loss of efficiency = ͑Actual LHs – Reasonable LHs͒ in impacted period ͑in LHs͒ Table summarizes the analysis The “reasonable labor-hour level” is 76.4% ͑14,117/18,482͒ In other words, the actual productivity was more than the planned productivity in February to June of 2008 Reasonable labor hours for base contract and change orders in the impacted period ͑July to October 2008͒ are the product of the earned hours in the same period and the “reasonable labor-hour level.” The inefficient labor hours due to the cumulative impact of change orders are 11,393 JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / AUGUST 2010 / 831 J Constr Eng Manage 2010.136:826-833 Table Inefficient Labor Hours with Method Used in Bell BCI Co v United States ͑2008͒ Table Lost Productivity for the Case Study Description Method Quantity Downloaded from ascelibrary.org by New York University on 05/10/15 Copyright ASCE For personal use only; all rights reserved Planned quantity performed in unimpacted period Planned labor hours for the first 17,000 units Actual labor hours for the first 17,000 units “Reasonable labor-hour level” Reasonable labor hours for base contract and change orders in impacted period Actual labor hours in impacted period Inefficient labor hours 17,000 18,482 14,117 76.4% 16,550 27,943 11,393 From The Appeal of P J Dick, Inc „2001… 6,276 6,858 6,532 11,393 3,976 6,138 Discussion This variant is used when the period without owner-caused disruptions is not available for the same work A similar work with an undisrupted period needs have to be identified in the same project or from a similar project However, productivities are not compared directly to find the loss of efficiency as in the measured mile analysis with a similar work Instead, an “efficient factor” is determined as the ratio of actual labor hours and budgeted labor hours for the similar work in the undisrupted period Realistic budgeted labor hours for the disrupted work are calculated by multiplying the budgeted labor hours with the “efficient factor.” This analysis is similar to the one used in Bell BCI Co v United States ͑2008͒ except the involvement of the similar work Therefore, similar formulas are as follows: Efficient factor = Earned value analysis Measured mile analysis Variant in James Corporation d/b/a James Construction v North Allegheny School District, et al ͑2007͒ Variant in Bell BCI Co v United States ͑2008͒ Variant in The Appeal of P J Dick, Inc ͑2001͒ Total cost method Lost productivity ͑labor hours͒ Actual LHs in unimpacted period for similar work Budgeted LHs in unimpacted period for similar work Realistic budgeted LHs = ͑Budgeted LHs in impacted period͒‫͑ء‬Efficient factor͒ Loss of efficiency = ͑Actual LHs – Realistic budgeted LHs͒ in impacted period ͑in LHs͒ For illustrative purposes, assume that the work during the disrupted period ͑July to October 2008͒ was different; however, it was similar to the work performed in February and March 2008 Table illustrates the analysis The efficient factor for the similar work is 1.106 The budgeted labor hours in the impacted period are the earned hours in the same period ͑Table 3͒ The inefficient labor hours are 3,976 Table Inefficient Labor Hours with Method Used in The Appeal of P J Dick, Inc ͑2001͒ Description Quantity Budgeted labor hours for similar work ͑February to March 08͒ Actual labor hours for similar work ͑February to March 08͒ “Efficient factor” Budgeted labor hours in the impacted period Realistic budgeted labor hours in the impacted period Actual labor hours in the impacted period Inefficient labor hours 2,648 2,929 1.106 21,667 23,966 27,943 3,976 The above case study illustrates that lost productivity can be proved and quantified by various analyzes These analyzes have been successfully applied in litigation cases for cumulative impact claims Though they all are acceptable to courts and boards, their results can be significantly different Table summarizes lost productivity estimated by these methods for the case study The inefficient labor hours based on the total cost method are the difference between the actual labor hours and the sum of planned and change order labor hours With the case study above, the loss of productivity is estimated in a range of 3,976 and 11,393 labor hours ͑Table 8͒ Therefore, the claimant should choose an appropriate analysis for a particular case based on the nature of the cumulative impact and the available project data Though the traditional measured mile analysis is prone to be the most credible method among those presented, it is not always applicable As previously discussed, both earned value and measured mile analyzes for determining the loss of productivity have their own limitations that may make them less useful in many cases For that reason, their combinations may be more effective and acceptable in legal proceedings in such circumstances In Bell BCI Co v United States ͑2008͒, the contractor had not achieved the originally planned productivity when extensive changes did not occur That is, during the undisrupted time the contractor had spent 1.08 h to earn one hour in the budget Thus, the calculation of inefficiency would have not been accepted if earned value analysis had been employed In The Appeal of P J Dick, Inc ͑2001͒, a traditional measured mile study could not be used because the measured mile period simply did not exist Additionally, the similar work ͑feeder circuit͒ was not too similar to the impacted work ͑branch circuit͒ to directly compare their productivity Therefore, the lost productivity was indirectly determined through the efficient factor of the similar work The combinations of the two techniques proved to work well in those situations A claimant may apply different approaches for comparison to obtain a more reliable result Contractors, owners, experts, and attorneys usually employ more than one method when assessing claims ͑Jones 2003͒ This practice is also applicable in lawsuits In The Appeal of P J Dick, Inc ͑2001͒, the contractor’s expert used the industry study ͑MCAA͒ as an alternate method to demonstrate the inefficiency of the electrical work in addition to a measured mile analysis Practically, at least two parties are involved in a dispute and an approach that works well for a party is not always in the best interest of the other In the case study, for example, while the contractor may prove its claim with the Bell BCI variation ͑11,393 labor-hour loss͒, the owner may disprove this claim with the Appeal of P J Dick variation ͑3,976 labor-hour loss͒ In terms 832 / JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / AUGUST 2010 J Constr Eng Manage 2010.136:826-833 Downloaded from ascelibrary.org by New York University on 05/10/15 Copyright ASCE For personal use only; all rights reserved of reliability, the descending orders of these approaches in general are measured mile analysis, Bell BCI variation, the Appeal of P J Dick variation, James Corp variation, and earned value analysis Measured mile analysis is most credible because it directly compares productivities of the same work between impacted and unimpacted periods Bell BCI variation and the Appeal of P J Dick variation employ the “reasonable labor-hour level” and “efficient factor,” respectively, to adjust budgeted labor hours However, the Appeal of P J Dick variation calculates the efficient factor based on the similar work not the same work as in Bell BCI variation James Corp variation calculates inefficiency by extrapolating actual labor hours in the unimpacted period to impacted period for the whole contract work and not for a specific work being disrupted due to multiple changes Earned value analysis is based on the original estimate which is not always reasonable Therefore, unless a contract specifies otherwise, results from analysis of a more reliable approach should be chosen and/or should be a basis for negotiation if available data allow conducting such analysis That is, the results of measured mile analysis are preferred to those of Bell BCI variation which in turns are preferred to those of the Appeal of P J Dick variation and so on This may avoid any disagreement for the wide range of results drawn from different approaches Conclusions Project specific studies such as measured mile analysis, earned value analysis, and their variants can demonstrate two ͑causation and resultant injury͒ of the three elements of successful lost productivity claims They have been the most widely accepted by legal bodies The variants which combine the earned value and measured mile methods are derived from recent law suits for proving and quantifying the cumulative impacts of change orders They can overcome some of limitations of measured mile and earned value analyzes These variants use the project-specific records to establish loss of efficiency unless a traditional measured mile study can be applicable Therefore, they are feasible alternatives for claimants and claim analysts to prove, quantify, and evaluate lost labor productivity When being applied to the same case, these methods generally yield significantly different results of loss of efficiency The illustrations and comparisons in the case study in this paper indicate that the amount of lost productivity depends a lot on how a measured mile analysis and its variants are used However, it is difficult to justify which method is more accurate and reliable in a particular case They should be selected based on the availability of project records and the nature of changes and cumulative impacts Practitioners can also employ two or more methods to enhance the credibility of their lost productivity analysis References Amelco Electric v City of Thousand Oaks, 27 Cal 4th 228, 115 Cal Rptr 2d 900, 38 P.3d 1120, rehearing denied by 2002 Cal LEXIS 1689 ͑March 13, 2002͒ The Appeal of Coates Industrial Piping, Inc., VABCA No 5412 ͑July 26, 1999͒ The Appeal of Gulf Coast Trading Co., ENGBCA No 5795 ͑1994͒ The Appeal of P J Dick, Inc., VABCA Nos 6080–6082, 01-2 BCA ͑CCH͒ ͑September 27, 2001͒ Association for the Advancement of Cost Engineering ͑AACE͒ International, Inc ͑2004͒ “Estimating lost labor productivity in construction claims.” AACE International Recommended Practice No 25R-03, Morgantown, W.Va Atlas Constr Co., Inc., GSBCA No 8593, 90-1 B.C.A ͑CCH͒ ͑1990͒ Bell BCI Co v United States, No 03-1613C, 81 Fed Cl 617 ͑April 21, 2008͒ C Norman Peterson Co v Container Corp., 172 Cal App 3d 628, 218 Cal Rptr 592 ͑1985͒ Centex Bateson Construction Co., VABCA No 4613, 99-1 BCA 30,153, at 149,259 ͑1998͒ Central Mechanical Constr., ASBCA No 29434, 86-3 BCA ͑CCH͒ ͑1986͒ Charles G Williams Constr., Inc., ASBCA No 33766, 89-2 B.C.A, CCH ͑1989͒ Clark Concrete Contractors, Inc v General Services Administration, GSBCA No 14340, 99-1 BCA ͑CCH͒ ͑1999͒ The Clark Construction Group, Inc., VABCA No 5674, 00-1 B.C.A ͑April 5, 2000͒ Coley Properties Corp v United States, 593 F.2d 380 ͑1979͒ David J Tiernay, GSBCA Nos 7107, 6198, 88-2 BCA ͑CCH͒ ͑1988͒ Dawson Constr Co., VABCA Nos 3306–3310, 93-3 BCA ͑CCH͒ ͑1993͒, affirmed, 34 F.3d 1080 ͑Fed Cir 1994͒ Dyson & Co., ASBCA No 21673, 78-2 BCA, CCH ͑1978͒ Electronics & Missile Facilities v United States, 416 F.2d 1345 ͑Ct Cl 1969͒ Freeman-Darling, Inc., GSBCA No 7112 ͑April 27, 1989͒ Fru-Con Constr Corp v United States, 43 Fed Cl 306 ͑Ct Cl 1999͒ Gulezian, R., and Samelian, F ͑2003͒ “Baseline determination in construction labor productivity-loss claims.” J Manage Eng., 19͑4͒, 160–165 Hanna, A S ͑2001͒ “Quantifying the cumulative impact of change orders or electrical and mechanical contractors.” Research Report 158– 11, Construction Industry Institute, Univ of Texas at Austin, Austin, Tex Ibbs, C W., and Allen, W E ͑1995͒ “Quantitative impacts of project change.” Source Document 108, Construction Industry Institute, Univ of Texas at Austin, Austin, Tex Ibbs, W., and Liu, M ͑2005͒ “Improved measured mile analysis technique.” J Constr Eng Manage., 131͑12͒, 1249–1256 Ibbs, W., Nguyen, L D., and Lee, S ͑2007͒ “Quantified impacts of project change.” J Profl Issues Eng Educ Pract., 133͑1͒, 45–52 James Corporation d/b/a James Construction v North Allegheny School District, et al., No 1268 C.D 2007 ͑Pa Comwlth November 30, 2007͒ Jones, R M ͑2003͒ “Update on proving and pricing inefficiency claims.” Constr Lawyer, 23͑Summer͒, 3–11 Leonard, C ͑1988͒ “The effect of change orders on productivity.” MS thesis, Concordia Univ., Montreal, Quebec Litton Systems, Inc., Ingalls Shipbuilding Division, ASBCA No 17578, 78-1 B.C.A, CCH ͑1978͒ Loulakis, M C., and Santiago, S J ͑1999͒ “Getting the most out of your ‘measured mile’ approach.” Civ Eng (N.Y.), 69͑11͒, 69 McConnell, H H ͑1997͒ “Distinguishing quantum meruit and unjust enrichment in the construction setting.” The Florida Bar Journal, 71͑3͒, 88 Pittman Constr Co v United States, Cl Ct 211 ͑1983͒ Rice v United States, 317 U.S 61, 64-65 ͑1942͒ Southwest Marine, Inc., DOT BCA No 1663, 94-3 BCA, CCH ͑1994͒ State ex rel Dept of Transp v Guy F Atkinson Co., 187 Cal App 3d 25, 231 Cal Rptr 382 ͑1986͒ Thomas, H R., and Sanvido, V E ͑2000͒ “Quantification of losses caused by labor inefficiencies: Where is the elusive measured mile?” Constr Law Bus., 1͑3͒, 1–14 Thomas, H R., and Završki, I ͑1999͒ “Construction baseline productivity: Theory and practice.” J Constr Eng Manage., 125͑5͒, 295–303 Triax Co v United States, 28 Fed Cl 733 ͑1993͒ Watt Plumbing, Air Conditioning & Elec., Inc v Tulsa Rig, Reel & Mfg Co., 533 P.2d 980, 983 ͑Sup Ct Okla 1975͒ Zink, D A ͑1986͒ “The measured mile: Proving construction inefficiency costs.” Coastal Eng., 28͑4͒, 19–21 JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT © ASCE / AUGUST 2010 / 833 J Constr Eng Manage 2010.136:826-833  ... for example, the combinations of earned value and measured mile analyzes, have successfully been used in cumulative impact claims Cases with Cumulative Impact Claims Courts and boards of contract... representative cases that courts or boards denied and allowed cumulative impact claims, respectively The cases in each table are listed chronologically In Table cumulative impact claims were denied... ͑September 27, 2001͒ Association for the Advancement of Cost Engineering ͑AACE͒ International, Inc ͑2004͒ “Estimating lost labor productivity in construction claims. ” AACE International Recommended