Feb 28, 1994 - These types of projects create a serious disruption in the ... state highway agencies throughout the United States. In the past few years, several innovative contracting methods have been ..... Maryland. 3,200. 35,087,606. 571.
TIME Is MONEY: INNOVATIVE CONTRACTING METHODS IN HIGHWAY CONSTRUCTION
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
By Zohar J. Herbsman,· Member, ASCE, Wei Tong Chen,2 and William C. Epstein,3 Student Member, ASCE ABSTRACT: In today's highway construction industry, the focus has shifted from building new transportation facilities to resurfacing, rehabilitating and restoring (3R's) those already in existence. These types of projects create a serious disruption in the normal flow of traffic resulting in a major inconvenience to the traveling public. Reducing construction time, thereby mitigating this inconvenience, has become a major emphasis for state highway agencies throughout the United States. In the past few years, several innovative contracting methods have been introduced into the highway construction industry. The basic concept of these innovative techniques is the motivation they provide encouraging the contractor to work faster, schedule accurately, and manage the construction process better. Results from the research conducted by the writers indicate that recent transportation projects constructed utilizing these innovative contracting methods have been very successful in reducing construction time, while only increasing overall construction costs minimally, if at all. This paper examines the advantages and disadvantages of these methods analyzing data obtained from various case studies throughout the United States and abroad.
INTRODUCTION
Contract time, which can be defined as the maximum time allowed for the contractor to complete all work as specified in the contract documents, is one of the most important aspects of the entire construction process. Not only do these time calculations affect areas such as budgeting, resource planning, local economies, and claims issues, but reasonable contract time may avoid higher bid costs as well as decrease the possibility of disputes between the contractor and the contracting authority (Hancher and Rowings 1981). Today's highways have become such an integral part of our daily lives that over 90% of America's total transportation needs are met by these systems (Highway 1992). Unfortunately, this heavy traffic has had an undesirable effect of decreasing the ability of each motorist to effectively utilize this country's transportation network. In the past two decades, the number of vehicles on our roadways has increased by more than 75% while the total mileage of new highway systems has increased by only 4% (Highway 1992). As a result, the traveling public inevitably faces heavy traffic congestion, especially in the metropolitan areas. Therefore, increasing usable roadway lanes for vehicular traffic has become one of the top priorities of all state highway agencies (SHAs) across the nation. To effectively meet the growing need for additional usable highways, attention is shifting from the building of new transportation facilities to the resurfacing, rehabilitation and restoration (3R's) of those already in existence. Typically, these projects are undertaken in heavily urbanized areas, causing extreme traffic congestion during the construction period. This slowdown of the traffic flow not only poses severe inconvenience to the traveling public, it also negatively impacts the business community. With such a premium placed on 'Prof., Dept. ofCiv. Engrg., Univ. of Florida, 345 Weil Hall, Gainesville, FL 32611. 'Assoc. Prof. of Constr. Engrg., Dept. of Arch. and Urban Des., Chung Hua Polytech. Inst., Hsin Chu, Taiwan. 'PhD Candidate, Dept. of Civ. Engrg., Univ. of Florida, 345 Weil Hall, Gainesville, FL. Note. Discussion open until February I, 1996. To extend the closing date one month, a written request must be filed with the ASCE Manager of Journals. The manuscript for this paper was submitted for review and possible publication on June 16, 1994. This paper is part of the Journal o/Construction Engineering and Management, Vol. 121, No.3, September, 1995. ©ASCE, ISSN 0733-9364/95/0003-0273-0281/$2.00 + $.25 per page. Paper No. 8672.
construction time, many of this country's SHAs have begun to implement alternate construction procedures in an attempt to reduce project durations. A variety of contracting and bidding methods have been introduced in the United States over the last several years that specifically address reduction of construction times. The four most popular of these methods presently being utilized are: • Bidding on cost/time • Incentive/disincentive (I/O) • Bidding on cost/time combined with incentive/disincentive • Lane rental Each of these four innovative methods considers the value of time. Due to the significant influence of time value, schedulers will have to become much more proficient in estimating construction times. It is therefore a necessity for the successful practitioner to become familiar with the basic principles and procedures of these developing contracting methods. TIME VALUE
Each of the innovative methods addressed in this paper is based on the principle of cost reimbursement to the contractor for contract time reduction. To ascertain these costs, the owner has to calculate what is the value of the time. For example, if a contractor reduces construction time by 40 calendar days, the owner has to determine what is the value of the 40 calendar days of time reduction. Calculation of the time value is based on a parameter defined as "unit time value" (UTV), which represents the value of a time unit to the owner. The measurement of the UTV can be in any unit that the owner chooses, such as hours, days, weeks or months. The total value of the time is calculated using the following formula: TTV = UTV x T
(1)
where TTV = total time value; UTV = unit time value; and T = time. For example, the owner determines that reduction of the contract time is valued at $4,000/day (UTV), and if the contractor reduces construction time by 20 days (D, then the total time value will be TTV = $4,000/day x 20 days = $80,000.
JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT / SEPTEMBER 1995/273
J. Constr. Eng. Manage., 1995, 121(3): 273-281
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
CALCULATION OF UNIT TIME VALUE
The UTV represents the cost of delays to the owner. In most cases, the UTV will include the direct cost resulting from construction delays, such as temporary facilities, moving costs, and other alternate solutions. Indirect cost items encompassing both job overhead and general overhead can also be considered in the UTV calculation. A variety of other general costs involving losses to the business community, reduction of potential profits and even hardship to the owner, though harder to quantify, can also be incorporated into the final calculations. Computation of the UTV can be demonstrated by two case studies. The first case study involves the renovation of a small office complex. In this project the owner has to relocate his business to a new location temporarily. To calculate the UTV, the owner considers the expense associated with renting these temporary facilities, additional transportation costs, as well as advertisement of the new office location. A UTV of $3,000/ week was considered sufficient to cover the direct costs due to the building renovation. This UTV amount however, does not include any indirect costs such as the inconvenience to the business, loss of revenue, or other related items. The second case study concerns the construction of a major hotel in central Florida. The UTV was estimated at $25,000/ day consisting of both direct and indirect costs. Indirect cost items included the loss of profits and potential clients, as well as other cash-flow problems that could arise due to a delay in the final completion date. CALCULATION OF DAILY ROAD-USER COST
UTV calculations in the highway construction industry are typically expressed as a cost per day. This variable, evaluated by SHAs throughout the country, is commonly referred to as the "daily road-user cost" (DRUC). For example, if a new road has to be built, a feasibility study will be conducted by the SHA to determine the economic impact of the new facility. A calculation of the DRUC will be part of this analysis, and will include such items as travel time, travel distance, fuel expense, as well as other related components. Appendix I is an example of a DRUC calculation as used by the Kansas Department of Transportation (DOT) to ascertain the daily cost associated with a 33 km (21 mi) detour. Although DRUC calculations have been performed routinely by SHAs for many years, no formal computational procedures have been instituted nationwide. Therefore, many states have developed their own procedures. One of the major differences in DRUC calculations among the various SHAs is the way in which they approach indirect costs. Many states, in fact, due to regulatory constraints and hard-to-quantify values, do not even consider indirect costs in their DRUC calculations. This is a critical issue because the indirect cost items can be very significant, especially for projects constructed in heavily populated areas. For instance, on a major highway reconstruction project resulting from the earthquake damage in California, the DRUC was estimated at $200,000/ day based on direct costs only. Preliminary evaluations by California DOT indicated that had they included indirect costs, this figure could have been increased substantially. Since UTV and DRUC calculations are the basis for determining time value in all innovative methods, the importance of the establishment of standardized procedures for these calculations cannot be stressed enough. INNOVATIVE CONTRACTING METHODS Overview
The utilization of innovative contracting methods has increased substantially in recent years. The evaluation of these
nontraditional techniques has been performed by the authors researching dozens of case studies of projects constructed under one of these contracting strategies. To be able to compare and contrast these methods, the reader needs to be familiar with the basic principles of each method. This section is devoted to describing concepts and procedures associated with each of the four most popular innovative contracting methods used in today's highway construction industry. Bidding on CostlTime (A + B)
Until recently, bidding on cost/time, also referred to as the + B method, has been employed very sparingly in the United States. However, after this approach was recommended by the Federal Highway Agency, its use has steadily increased among various SHAs. The first step in this procedure is for the owner to establish the DRUC and incorporate it into the bid documents. Next, every contractor who participates in the bidding process will be required to calculate two values: A-the estimated construction cost for the project; and B-the estimated project duration for project completion. The successful bidder is the contractor who submits the lowest total combined bid (TCB) using the following formula: A
TCB
=
ECC + (DRUC x EPD)
(2)
where TCB = total combined bid; ECC = estimated construction cost for the project (the "A" in the A + B method); DRUC = daily road-user cost; and EPD = estimated project duration for project completion (the" B" in the A + B method). The calculations of the TCB can be illustrated using an example of a highway bid tabulation from a construction project in North Carolina (Table 1). Since this project was contracted utilizing the A + B method, each contractor participating in the bid submitted an estimated construction cost, column 2, and the estimated project duration required to complete this project, column 4. Table 1 illustrates that the estimated construction costs submitted by the five bidding contractors range from $19,371,550 to $21,138,086, while the estimated project durations vary from a low of 642 days to a high of 762 days. Although contractor B ranked second in both the estimated construction cost and the estimated project duration, this contractor was awarded the project by virtue of the lowest total combined bid of $24,222,537. The calculation of the total combined bid of contractor B is as follows: $19,518,537 + 672 days x $7,000/day = $24,222,537. Upon further analysis of the bid results from Table 1, one can argue that the North Carolina DOT overpaid $146,987, which is the difference between the lowest estimated construction cost of contractor A ($19,371,550) compared to that of contractor B ($19,518,537), by selecting contractor B over contractor A. However, from the general public's point of view, the state of North Carolina obtained a project with a 90-day reduction in construction time, which is calculated by subtracting the estimated project duration of contractor B at 672 days from the engineer's time estimate of 762 days (762 days - 672 days = 90 days). According to the North Carolina DOT's DRUC calculations, this savings of 90 days is valued at $630,000 (90 days x $7,000/day). Therefore, the net savings realized by the citizens of North Carolina utilizing the A + B method was $483,013 ($630,000 - $146,987). It should be noted that the savings to the general public will actually be much higher because the North Carolina DOT did not consider any indirect costs in their DRUC computations.
274/ JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT / SEPTEMBER 1995
J. Constr. Eng. Manage., 1995, 121(3): 273-281
TABLE 1. Bid Results Utilizing the Bidding on CostITlme (A Feb 28, 1994)
+
B) Method (G. C. Gibson, personal communication, North Carolina DOT, Total Time Value
Estimated Construction Cost
Contractor
Dollar value (A)
Rank
(1 )
(2)
(3)
Estimated project duration (d) (4)
A B
19,371,550 19,518,537 19,734,919 20,198,158 21,138,086
1 2 3 4 5
762 672 702 642 762
C D
E
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
The feedback from those SHAs who have implemented the bidding on cost/time method has been very positive (Harp 1990; Tarricone 1993). Herbsman and Ellis (1992) analyzed data obtained from 14 case studies of projects contracted using the bidding on cost/time method. The results of this analysis indicated that for the most part, substantial time reduction was achieved with little or no increase in overall construction costs. Table 2 is a summary of those 14 case studies investigated by Herbsman and Ellis (1992). Table 2 illustrates that in 11 of 14 projects, the estimated project duration submitted by the successful bidder as shown in column 6 was less than the engineer's time estimate, column 5. The time savings realized in these 11 cases ranged from 2.9% to 68.4% with a 29.87% average reduction of project durations. In two instances, the estimated project duration submitted exceeded the engineer's estimate, and there was one project in which the contractor's estimate equaled the engineer's time estimate. Table 3 summarizes the results of 11 projects that were awarded in North Carolina during the period from 1991 through 1993 utilizing the A + B method. Fig. 1 shows this raw data on a histogram, indicating that in every case the estimated project durations submitted by the successful bidders were less than or equal to the North Carolina DOT engineer's time estimate. The time savings realized in these 11 cases ranged from 0.00% to 62.26% with a 31.56% average reduction of TABLE 2 Case study (1 ) 1 2 3 4 5 6 7 8 9 10 11 12
13 14
State (2)
(3)
Project cost ($) (4)
5,000 5,000 7,000 5,000 5,000 5,000 5,000 5,000 3,200 7,000 20,000 5,000 5,000 5,000
3,034,765 2,306,380 1,361,009 16,329,262 12,583,349 9,186,877 18,554,123 15,636,180 35,087,606 4,271,599 1,637,015 39,833,648 39,781,121 15,867,833
Delaware Delaware Georgia Kentucky Kentucky Kentucky Kentucky Kentucky Maryland Mississippi Missouri Texas Texas Texas
B)
(B) (6)
Dollar value
Rank
(5)
(7)
(8)
4 2 3 1 4
5,334,000 4,704,000 4,914,000 4,494,000 5,334,000
24,705,550 24,222,537 24,648,919 24,692,158 26,472,086
4 1" 2 3 5
=
762 d.
estimated project durations. Furthermore, it should be noted that to date, 5 of these 11 projects have been completed, and of these 5, all were finished on or before the contractor's estimated completion date. The bidding on cost/time method has been gaining acceptance among many of this country's SHAs. A list of 16 states that currently use this method is presented in Table 4. Of the SHAs surveyed that have contracted projects under the A + B method, only the Texas DOT has met with any adverse consequences. Incentive/Disincentive The incentive/disincentive (I/O) approach of motivating general contractors to reduce overall contract durations is probably the most established of the four innovative contracting methods being analyzed in this paper. In this method, the contract time is determined by the owner and presented as part of the bid documents. If the contractor is able to complete the project ahead of schedule, this contractor would then be entitled to a bonus (incentive fee). If, on the other hand, the contractor finishes the project behind schedule, a penalty (disincentive fee) is then assessed by the owner. Although the basic principles of utilizing the I/O method are similar within most transportation departments, there are a number of procedural variations. Calculations of Daily I/O Fees As was the case with the bidding on cost/time (A + B) method, the ORUC is also an integral part of the I/O method.
Summary of Ca. .Study Results of Projects Contracted under A DRUC ($/day)
+
Dollar value" Rank
Note: engineer's cost estimate (A) = $20,568,042; engmeer's ltme estImate "Equals column 4 x DRUC ($7,OOO/d). "The successful bidder.
Bidding on Cost/Time-Commentary on Current SHA Applications
Total Combined Bid (A
+ B Method (Herbsman and Ellis 1992)
Engineer's time estimate (d)
Contractor time bid (d)
Time saving" (d)
(5)
(6)
(7)
(%) (8)
170 160 90 1,094 153 474 643 729 571 200 30 1,040 960 75
120 125 111 517 90 150 500 450 571 151 53 1,010 900 61
50 35 -21' 577 63 324 143 219 0 49 -23' 30 60 11
29.4 28.0 -23.3" 52.7 41.2 68.4 22.2 38.3 0.0 24.5 -76.7' 2.9 6.3 14.7
Time saving b
"Column 5 - column 6. "(Column 71 column 5) x 100. 'Number that the contract time exceeds the engineer's time estimate. JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT 1SEPTEMBER 1995/275
J. Constr. Eng. Manage., 1995, 121(3): 273-281
TABLE 3. Results of A + B Projects Awarded by North Carolina DOT (G. C. Gibson, North Carolina DOT, personal communication, Feb. 28, 1994) Project no. (1)
Engineer's time estimate (2)
Estimated project durationa (3)
580 226 793 762 636 763 552 856 275 674 835
I
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
2 3 4 5 6 7 8 9
10 11
Difference Daysb (4)
Percent" (5)
Time value in $/day (6)
305 215
275
600
193
672 240 427 306 800 170 674 571
396 336 246 56 105 0 264
47.41 4.86 24.34 11.81 62.26 44.03 44.57 6.54 38.18 0.00 31.62
1,000 2,000 10,000 7,000 2,000 10,000 7,000 2,000 7,000 12,000 3,100
11 90
"Time bid by the successful bidder. "Value in column 2 minus value in column 3. 25,000
11.4 54.4 25.7 5.7 2.8
TABLE 6. Schedule of Dally I/O Fees UtiliZed by New Jersey DOT (Robert K. Suess, New Jersey DOT, personal communication, Aug.
17,1992) Total project cost (in millions of dollars)
Daily I/O rates
(1 )
(2)
0-0.5 0.5-1.5 1.5-5.0 5.0-10.0 10.0-15.00 15.0-20.0 20.0-30.0 30.0-40.0 40.0-50.0 50.0+
$1,000 $2,000 $5,000 $6,000 $8,000 $10,000 $13,000 $16,000 $17,000 0.03% of total project cost
700
i;'
C
600
.5 c
500
0
!= 400 C
TABLE 5. Dally I/O Fees Used by SHAs (Roger E. Bierbaum, Iowa DOT, personal communication, Oct. 9,1991)
300
200 100 1
2
3
4
5
6
7
8
9
10
11
Project No. _ FIG. 1.
Engineer's Estimate _
Results of A
Low Bid Estimate
+ B Projects Awarded by North Carolina DOT
(1991-93)
TABLE 4. The "A + B" Contracting Project Status In United States (Alan Rockne, FHWA, personal communication, Dec. 9,1994) location (1 )
Arkansas California Delaware District of Columbia Idaho Kentucky Maryland Michigan Mississippi Missouri Nebraska North Carolina Pennsylvania Texas Utah Virginia
Remarks (2)
One project, favorable results Three projects, no results to report Three projects, first two had favorable results, third had problem unrelated A + B bidding Two projects, favorable results One project, no results to report One project, favorable results Several projects, favorable results Two projects, too early for results Two projects, favorable results Numerous projects with favorable results One project, favorable results Ten projects, generally favorable results Two projects, favorable results Two projects, unfavorable results, A + B created friction between the State and the contractor Three projects, favorable results One project, no results to report
Most SHAs calculate the DRUC and apply this value as their daily incentive/disincentive fee. Table 5, based on a study conducted by the Iowa DOT, illustrates the distribution of I/O fees for various transportation agencies across the United States. According to the SHAs surveyed, over half of all I/O contracts let by these agencies had an I/O fee in the range of $2,500/day to $5,OOO/day. It should also be noted that, by far, the most popular fee used was $5,OOO/day. The data of Table 5 indicate that at the time of the survey, only 2.8% of the I/O contracts were mandating fees in excess of $25,000/day. However, as the trend for more urban highway rehabilitation continues, the magnitudes of the I/O fees are expected to increase profoundly as greater significance is placed on the value of time. Already, there are a few rehabilitation projects under way in major metropolitan areas, where DRUC calculations have yielded values in the range of hundreds of thousands of dollars per day. In one recent California emergency transportation project, DRUC estimates were used as the basis for contractual establishment of a $200,000/day liD fee. The contractor on this particular project finished 73 days ahead of schedule and received 14.6 million dollars in incentive fees. Although most states are using the DRUC as a basis for determining appropriate liD fees, there are some states using different parameters for establishing I/O fees. One of the alternate approaches is to calculate the I/O fee as a percentage of total project cost. Table 6 is a schedule of I/O fees as utilized by the New Jersey DOT. For example, projects in which total construction costs are expected to be between $1,500,000 and $5,000,000, the I/O fees will be set at $5,000/ day (column 2). For projects whose costs are to exceed $500,000,000 the lID fees will be 0.03% of the total project cost.
Relationships between Incentive and Disincentive Fees Most states that are using the liD method apply the same value for both the incentive and disincentive fee. Table 7, based on the Iowa DOT study, indicates that of the 35 states surveyed using the I/O method, 28 (80%) had an I-fee equal to the D-fee (l = D). Four states (11.4%) said that no sys-
276/ JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT / SEPTEMBER 1995
J. Constr. Eng. Manage., 1995, 121(3): 273-281
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
TABLE 7. 110 Fees Used by Various SHAs (Roger E. Bierbaum, Iowa DOT, personal communication, Oct. 9, 1991) No.
State
Relationship of I fees to 0 fees·
(1)
(2)
(3)
I 2 3 4 5 6 7 8 9 10 II
Alabama Arizona Arkansas California Colorado Delaware Florida Georgia Idaho Illinois Indiana Iowa Kansas Maine Maryland Massachusetts Michigan Minnesota Missouri Montana Nevada New Hampshire New Jersey New York North Carolina North Dakota Ohio Pennsylvania South Dakota Tennessee Utah Virginia Washington Wisconsin Wyoming
1= D 1= D 1= D 1= D I> D Varies 1= D Varies D = 0 1= D 1= D 1= D 1= D 1= D 1= D 1= D 1= D 1= D 1< D 1= D Varies 1= D I> D 1= D 1= D 1= D 1= D 1= D 1= D 1= D 1= D Varies 1= D 1= D 1= D
12 13
14 15 16 17 18 19 20 21 22
23 24 25 26 27 28 29 30 31 32 33
34 35
"I = incentive; D = disincentive.
tematic relationship existed between 110 fees and, they varied from project to project. Two states (5.7%) reported that their I-fee was higher than their D-fee (I > D), while one state (2.9%) had an I-fee smaller than its D-fee (I < D). One state (2.9%) indicated that they do not even assess aD-fee. Maximum I/O Fees Allowed Many states have implemented limits to the maximum amount of incentive fees that will be paid out and the disincentive fees that will be assessed to the contractor. Table 8, from the Iowa DOT study, illustrates several different approaches for instituting some type of limit on 110 fees. These limits, also known as caps, have been set by some states as a percentage of the total construction cost. According to Table 8 the maximum percentage cap reported was 10%, while the majority of those SHAs that limited 110 fees by means of a percentage cap used a value of 5% of the total construction cost. Other states are restricting the maximum amount of I/O fees to a flat-rate dollar amount. For example, New Jersey DOT has mandated that total I/O fees on anyone project cannot exceed $100,000. Only one state surveyed, Arizona, had a limiting cap in terms of time rather than dollars (± 30 days). Several of the SHAs that responded varied their cap amounts depending on the project. At the time of the survey, some states did not have any restrictions on the cap rates, but most indicated that they were considering some sort of limitations for the future.
TABLE 8. 110 Cap Rates for Various SHAa (Roger E. Bierbaum, Iowa DOT, personal communication, Oct. 9, 1991) No.
State
(1 )
(2)
(3)
Alabama Arizona Arkansas California Colorado Delaware Florida Georgia Idaho Illinois Indiana Iowa Kansas Maine Maryland Massachusetts Michigan Minnesota Missouri Montana Nevada New Hampshire New Jersey New York North Carolina North Dakota Ohio Pennsylvania South Dakota Tennessee Utah Virginia Washington Wisconsin Wyoming
None ±30 days None Dollar amount None None Varies None Varies NIA Dollar amount Dollar amount Dollar amount Dollar amount
1 2 3 4 5 6 7 8 9 10
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Cap
5% None
5% None
10% Varies Varies None
$100.000 10% Varies
5% 5% 5% None None Dollar amount Dollar amount
5% Varies
6-8%
Incentive/Disincentive-Commentary on Current SHA Applications In evaluating the success of the 110 contracting method from the perspective of the transportation agencies, each state was asked in what percentage of the projects let under this method were incentive fees eventually paid out to the contractor. Additionally, the Iowa DOT study attempted to quantify how much of the total incentive fees available were the contractors able to earn. Table 9 is a summary of the responses of each of the 35 states surveyed. From column 4, it can be seen, that 25 states (71.4%) reported that the contractor received some incentive fees on at least 3/4 of all projects that were contracted under the 110 method. Even in the three states where data was not available, SHA representatives acknowledged that the contractors were typically able to earn incentive fees. Only Massachusetts said that the contractor was regularly unable to obtain any incentive fees. According to the statistics of the Massachusetts DOT, the department only paid out incentives one out of every five jobs. When asked to report what percentage of the total incentive fees available were the contractors able to earn, the responses were more diverse. The percentages listed in column 5 revealed that four states had paid out the maximum in incentive fees on 100% of their projects. Analysis of Table 9 demonstrates that contractors were almost always able to earn some sort of incentive fee, and most practitioners interviewed consider this data proof of the success of liD contracting strategy as a method for effectively reducing construction times. However, some transportation officials candidly pointed out that the incentive fees paid out were more a
JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT I SEPTEMBER 1995/2n
J. Constr. Eng. Manage., 1995, 121(3): 273-281
TABLE 9. Summary of I/O Fee Ratios for Various SHAs (Roger E. Bierbaum, Iowa DOT, personal communication, Oct. 9, 1991) State
(1)
(2)
(3)
(4)
1 2 3 4 5 6 7 8 9
Alabama Arizona Arkansas California Colorado Delaware Florida Georgia Idaho Illinois Indiana Iowa Kansas Maine Maryland Massachusetts Michigan Minnesota Missouri Montana Nevada New Hampshire New Jersey New York North Carolina North Dakota Ohio Pennsylvania South Dakota Tennessee Utah Virginia Washington Wisconsin Wyoming
65% 100% 100% 100% 100% 100% 70% N/A N/A 90% 100% 90% 50% 100% 20% Most 98% Almost all 80% 100% Most 100% 100% 80% 95% 100%
25% 100% N/A 90% 100% 70% 40% N/A
75% 50% 25% 0% Few 50% 100% 20% 40% 50%
N/A
N/A
10
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
Ratio of I given Ratio of the I to for I/O projects the maximum I
No.
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
100% 75% 90% 90% 46% 80% 75% 90-\{)O%
N/A 67% 30% 80% 50% 100% 40%
N/A
50%
N/A 60% 30-40% 57% 50% 10% 60-80%
Source: Iowa DOT Survey, 1991.
function of conservative original engineers' time estimates, rather than a result of any special effort on the part of general contractors.
Bidding on Cost/Time Combined with Incentive/Disincentive Bidding on cost/time combined with the incentive/disincentive method is a composite technique commonly referred to as A + B plus 110. The low bidder is determined from the results of the A + B bid procedure. Upon selection of the successful bidder, the estimated project duration submitted by that contractor becomes the contractual project duration. Incentive/disincentive fees are then paid or collected with respect to this newly established contract duration. To demonstrate this method, data from a project that was constructed in Kentucky will be used. The successful bidder estimated the project duration at 450 days, which then became the official contract duration. The contractor completed this project in 406 days, 44 days ahead of schedule. The original contract documents had specified an 110 fee of $5,000/ day. Upon completion of the project the contractor therefore was entitled to an incentive fee of $220,000 (44 days x $5,0001 day). Table 10 lists 10 recent contracts let by Missouri DOT utilizing this combined method. Clearly, it can be seen that all 10 projects were completed, not only in less time than that allotted for in the initial engineer's estimates, but also faster than the successful bidders' original estimated project durations.
A + B plus I/O-Commentary on Current SHA Applications
To date, the number of projects let under this combined method is still relatively small. However, many of the SHAs interviewed were considering utilization of this combined approach in the near future in an attempt to further encourage reduction of construction times. The motivational factors provided to the general contractor under this method are twofold. Initially, through competitive A + B bidding, the contractor's estimated project durations are reduced with respect to the original engineer's time estimates. Moreover, after award of the contract, the successful contractor has an extra impetus to continue to reduce construction times in an effort to earn additional incentive fees. A number of practitioners expressed some reservations concerning the possibility of general contractors, who are confident that they will be the low bidder, purposely inflating their estimated project durations in order to realize substantial incentive fees upon early completion of the project. This problem, however, can be overcome by placing a stipulation in the contract documents stating that if the estimated project duration exceeds the engineer's time estimates the bid can be rejected.
Lane Rental Introduced in the United States in 1990, the lane-rental method has been used extensively by the British Department of Transportation (BDTp) since 1984. This method was defined as a way of providing financial incentives to general contractors and others to shorten the overall time required for lane closures (Report 1989). To utilize the lane-rental method, the transportation agency in charge of managing the project must set contract times, as well as determine the cost of lane closures under various working conditions. Each bidder is required to submit their cost estimate of the work to be performed, along with the amount of time needed for lane closures during the construction period. The total cost of the project is the sum of the cost estimate of work to be performed plus the cost of all essential lane closures. The lowest total aggregate cost estimate will determine who will be the successful bidder. Upon commencement of construction, whenever lane closures are required, the contractor will pay to the owner those charges in accordance with a predetermined schedule of lane closure fees. An example of a schedule of lane closure fees for a fourlane urban roadway construction project 2.4 km (1.5 mi) long is presented in Table 11. Referring to Table 11, if a contractor elects to close one lane of traffic from 9:00 a.m. to 3:00 p.m. on weekdays for a total of 20 days, a cost item of $120,000 ($6,000/day x 20 days) should be included in the bid. If on the other hand, this contractor decides to close one lane from 6:00 p.m. to 6:00 a.m. during weekdays for the same amount of days, then an add of only $20,000 ($I,OOO/day x 20 days) would be required in the bid. In order to avoid the payment of excessive lane closure fees, the contractor will strive to minimize construction work during the peak traffic hours. Bondar (1988) found that the contractors seem to better manage the construction work, as well as the material deliveries, thus maximizing productivity during lane closure periods, again in an effort to pay the lowest fees possible. Hence the contractor is highly motivated to minimize lane closures, and when such occurrences do arise, every effort is made to undertake these lane closures during off-peak hours. The net effect to the traveling public of this innovative contracting method is therefore a substantial reduction in the disruption to the normal traffic flow.
2781 JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT 1 SEPTEMBER 1995
J. Constr. Eng. Manage., 1995, 121(3): 273-281
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
TABLE 10. Missouri DOT Project Summary Utilizing "A + B" Plus 110 Method (Gary Chulllno, Missouri DOT, personal communication, April 7, 1993) Average daily traffic (vehiclesl day)
Project (1 )
Work type (2)
I 2 3 4 5 6 7 8 9 10
Lane reduction Bridge deck rehabilitation Bridge widening Interchange bridge replacement Pavement removal and replacement New interchange bridge Bridge replacement Two bridge replacement Bridge replacement Box culvert replacement
Note: DRUC = daily road-user cost: 1 "Maximum 10% of contract.
=
(3)
(4)
128,000 115,000 39,000 28,000 18,000 5,000 4,900 2,800 1,900 630
5,479,852 5,845,490 7,162,765 1,270,840 1,267,080 614,987 664,101 1,480,219 457,575 76,747
incentive; and D
=
Work duration
Weekdays Weekdays Weekends
9 a.m. to 3 p.m. 6 p.m. to 6 a.m. Friday 6 p.m. to Monday 6 a.m. Friday 6 p.m. to Monday 6 a.m.
Weekends
(2)
DRUC ($/unit) (5
Bid time
4,000/day
150 days 1,416 hr 65 days 136h 96 h 90 days 90 days 120 days 57 days 10 days
1,347/h 3,530/day 8,440/h
1,270/h 3,000/day 2,9OO/day 3,5OO/day 1,320/day
1,0lO/day
(6)
Completion time (7)
120 989 39 111 37 70 72 78 50 5
days hr days h h days days days days days
110 rates in dollars
(8)
120,000 (I) 584,549" (I) 91,780 (I) 123,084" (I) 74,930 (I) 60,000 (I) 52,200 (I) 147,000 (I) 9,400 (I) 5,050 (I)
disincentive.
TABLE 11. SChedule of Lane Closure Fees ("Lane Rental" 1990) Work time (1 )
Contract budget in dollars
Lane-closure fee (3)
$6,OOO/day for each lane closed
$1,OOO/day for each lane closed $10,000 for each lane closed $50,000 for whole project detour
Lane Rental-Commentary on SHA Applications
Although the lane-rental method has been utilized on a few projects in the United States (Colorado and North Carolina), this contracting strategy is still in its infancy. Due to the continued rate of success that this approach enjoys in the United Kingdom, some practitioners feel that the lane-rental method may become an established part of our highway construction industry. Maggs (1985) stated that "Lane-rental contracts for motorway reconstruction have been successful in demonstrating the scope for reducing the time needed for completion." According to Bondar (1988), lane-rental favors the more efficient firms who are able to provide careful thought for work planning. Analysis of the cost benefit of projects that were contracted using the lane-rental method shows these cost savings to be very significant. As a case in point, the BDTp has reported that during the first three years of using the lane-rental method, the equivalent dollar value of cost savings due to reduction in delays was worth approximately $40,000,000 (1986 value). Bondar (1988) extrapolated several conclusions from the experiences of the BDTp that can be applied to the U.S. highway construction industry. These conclusions can be divided into the following three major areas of interest: 1. Workload-The workload on the agency staff (resident engineers, inspectors, as well as other key employees) is considerably heavier compared to similar projects contracted under more conventional methods. A much greater emphasis must be placed on choosing the right personnel for the right shifts so as to maintain good communication with the contractors and their staffs at all times. 2. Lane Rental Fees-British contractors seemed to be much more motivated and overall better organized when operating under this type of contracting arrangement. It was also noted, as would be expected, that when contracting work under the lane-rental method, project scheduling was much tighter, and new and innovative construction techniques were
introduced trying to further minimize contract durations. A general contractor's sense of motivation seemed to be especially acute when the rental fees were high, considering a fee equivalent to $3,OOO/day or more to be worth the extra effort on their part. 3. Quality of work-According to the BDTp, the quality of work on projects that were constructed under the lanerental method was better than that of work produced by means of conventional methods, because the contractors were well aware of the penalties associated with delays due to nonacceptance of substandard work. The compressed time frames did adversely affect material supply schedules, however through better management the contractors were able to maintain their aggressive schedules. EVALUATION OF INNOVATIVE CONTRACTING METHODS
This paper presents four innovative contracting methods that all share a common goal of reducing project durations, but each has a very different approach for achieving this common objective. The writers have analyzed a great deal of data from case studies, as well as conducted numerous interviews with practitioners who have been involved in innovative contracting projects. The research results indicate that the four methods can be divided into two major areas. In the first category, time reduction is achieved by competition between bidding contractors. Of the four methods, both the bidding on cost/time (A + B) method as well as the A + B portion of the bidding on cost/time combined with incentive/disincentive (A + B plus liD) method are included in this category. In the second category, time reduction is encouraged by financial rewards and time delays are discouraged by penalties. The incentive/disincentive (liD) method, the liD portion of the bidding on cost/time combined with incentive/disincentive (A + B plus liD) method, and the lane-rental method are included in this category. The main conclusion reached from analysis of all available research data was that each of the four methods was successful in reducing construction time. This success rate was measured against engineers' time estimates calculated as if the projects were contracted under conventional methods. To properly evaluate which method provides the most effective time and cost benefits, the advantages and disadvantages of each will be analyzed.
JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT I SEPTEMBER 19951279
J. Constr. Eng. Manage., 1995, 121(3): 273-281
Bidding on CostlTime (A
+ B)
by 20% in recent years without witnessing any major delays in project completion dates.
Advantages
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
The most important advantage of this method is that time reduction is achieved through competition between contractors rather than direct monetary payments. A study of the bid results from a large number of projects has revealed that when bidding on cost/time, contractors do not appreciably raise their unit prices with respect to the unit prices of comparable projects bid using conventional methods. Even in those cases when the winning bid was not the lowest, in terms of direct construction costs, the public still benefitted by utilization of the A + B method with respect to construction time reductions. Disadvantages
A few of the highway industry personnel interviewed commented on potential problems that may arise under this method when contractors are too eager to get the job and may underestimate the construction time in order to increase their chances of being the successful bidder. Down the line, when such a contractor realizes that the original estimated project duration is unattainable, and disincentive fees are likely to be assessed, this contractor may be tempted to "cut corners" in an attempt to limit financial losses. At this point, the quality of the work may suffer, and the contractor will most likely be looking to change orders and claims as a means for recouping losses. The possibility of these antagonistic situations developing can be significantly reduced by providing specifications in the contract documents that allow the owner to reject an unreasonably low time estimate on the grounds of an unresponsive bid. Incentive/Disincentive (I/O)
Bidding on CostlTime Combined with Incentive/Disincentive
This method, still relatively untested in the United States, combines the advantages and disadvantages of each of the two techniques that make up this composite approach. To date, projects wherein the A + B plus I/O method has been utilized are those that are typically associated with an emergency-type situation, such as a major bridge repair. The main advantage associated with the combined method, as has been previously noted, is in the provision of a double motivational stimulus encouraging the contractor to reduce time. First, a low contract time is achieved by competition between contractors. Second, additional time reductions are achieved through incentive/disincentive fees. As more projects are constructed using this method, a more thorough evaluation can be prepared. Lane Rental
Advantages
The BOTp is pleased with the results they have had using the lane-rental method. The savings in cost and time have been substantial. According to the BOTp, the average reduction of construction times realized utilizing this method is approximately 25% (Srinivasan and Harris 1991). The main advantage of this method, is that it allows the contractors freedom to choose the best work patterns for construction (day, night, weekend, one lane closed, two lanes closed, detour, and so on). In addition, the contractors are strongly motivated to reduce the construction time because they actually payout (real money) to the owner for the lane closures.
Advantages
According to the survey conducted by the Iowa DOT, the utilization of I/O contracts achieved construction time reductions in almost every case. In fact, in over half of the projects studied, the contractor was able to achieve the maximum compensable time. Another advantage of the I/O contracting strategy is its flexibility in enabling SHAs to adjust their financial exposure by utilizing flat-rate or percentage caps for I/O fees. It has also been reported that on projects that used the I/O approach, as compared to the A + B method, the relationship between owner and contractor tended to be much less adversarial, which is apparently related to the fact that the contractor typically receives some amount of incentive payments. Disadvantages
The major disadvantage of the I/O method lies in the fact that the fees are based on the engineer's time estimates, as established by the SHAs. Today, in the United States, most transportation departments determine contract time based on the performance (production rates) ofthe average contractor. This practice of establishing contract times creates a situation whereby a good competitive contractor can reduce contract time with little or no additional commitment of resources. In other words, the same time reduction could essentially be achieved, free of charge, by reducing the original engineer's time estimates. This opinion is supported in several states (Florida and California, in particular) where, in recent years, original engineer's time estimates have been reduced, and contractors were still able to complete the projects on time. For example, the Florida DOT has reduced contract times
Disadvantages
The BOTp reported that the main disadvantage associated with the lane-rental method was the tendency for contractors to "cut corners" (similar to the previously noted disadvantages of the A + B method) when faced with the prospect of unanticipated lane-closure fees due to construction delays. Just as it is in the United States, this situation could lead to poor quality and strained contractor/owner relationships. CONCLUSIONS AND RECOMMENDATIONS
After careful analysis of the four innovative contracting methods presented in this paper, the writers have concluded the following: • All of the innovative methods have been successful in reducing construction times in almost every case. Time reductions of 20-50% were achieved in comparison to similar projects using conventional contracting methods. • The most economical method of the four is bidding on cost/time (A + B), since the time reduction is achieved through competition rather than monetary payments to the contractor. • The use of the incentive/disincentive method was determined to be less effective and more expensive than the A + B method. This would explain the decline in the number of I/O contracts let in the United States over the last few years. • The new approaches represented by the lane-rental method as well as the combined method (A + B plus I/O) are
280 I JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT / SEPTEMBER 1995
J. Constr. Eng. Manage., 1995, 121(3): 273-281
promising, however more data from additional case studies must be collected for further analysis. • More studies need to be done in developing standard procedures for calculating the value of time (DRUC), especially in the area of determining what indirect cost items have to be included in the final calculations. • Additional research should be conducted with the goal of establishing an acceptable methodology for computing "reasonable" contract durations. The application of knowledge-based exp~rt systems in this area has shown some promising results, and more research could yield valuable data.
Downloaded from ascelibrary.org by National Yunlin Univ on 03/08/16. Copyright ASCE. For personal use only; all rights reserved.
APPENDIX I. EXAMPLE OF DRUC CALCULATIONS USED BY KANSAS DOT
The following data are given (Procedure 1991): • Annual Average Daily Traffic (AADT): 20,000 vehicles/ day. • AADT for trucks: 1,600 vehicles/day. • AADT for cars: 18,400 vehicles/day. • Extra length travel cost for cars: $O.22/mi ($0.14/km). • Extra length travel cost for trucks: $0.711mi ($0.44/km). • Truckers wages plus fringe: $1O.50/h/vehicle. • Length of project affected by detour: 10 miles (16 km); speed limit 55 mph (88 km/h). • Length of detour: 21 mi (34 km); speed limit 55 mph (88 km/h). • Detour has 2 additional stops, one stop has a traffic signal. Direct Cost Calculation
Cost for cars is calculated as: (21 - 10) mi x 18,400 vehicles/day x $0.22/day
=
$44,528
Cost for trucks is calculated as: (21 - 10) mi x 1,600 vehicles/day x $0.711day Therefore, the total direct cost is: $44,528 + $12,496 = $57,024
=
$12,496
Time Difference Cost Calculation
10 mi divided by 55 mph = 0.1818 hr. 21 mi divided by 55 mph = 0.3812 hr. stop sign 2 min delay = 0.033 hr. traffic signal 2 min delay + 2 min delay to arrive at full speed = 0.0666 hr. time differential = 0.3812 hr + 0.033 hr + 0.066 hr 0.1818 hr time differential = 0.2984 hr. Cost = 0.2984 hr x 1,600 vehicles/day x $10.50/hr/vehicle = $5,013. 12/day. Daily Road User Cost
Daily Road User Cost (DRUC) = Direct Cost + Time Difference Cost DRUC = $57,024.00/day + 5,013.12/day DRUC = $62,037.12/day APPENDIX II.
REFERENCES
Bondar, V. A. (19R8). "Lane rental: the DTP view." J. Inst. of Highways and Transp., V(35), 22-26. Hancher, D. E., and Rowings, J. E. (1981). "Setting highway construction contract duration." J. Constr. Div., ASCE, 107(2), 169-179. Harp, D. W. (1990). "Innovative contracting practices-the new way to undertake public works projects." Hot Mix Asphalt Technol., Winter, 6-10. Herbsman, Z., and Ellis, R. (1992). "Multiparameter Bidding SystemInnovation in Contract Administration." J. Constr. Engrg. Mgmt., ASCE, 118(1), 142-150. Highway Statistics 1992. (1992). Federal Highway Administration, Washington, D.C. "Lane rental surfaces at ACPA convention." (1990). Highway and Heavy Constr., Mid-Jan., 9. Maggs, M. F. (1985). "Future trends in contracts and contract practices." J. Inst. of Highway and Transp., 9-13. Procedure for determining incentive/disincentive amount and calendar completion date. (1991). Kansas DOT, Topeka, Kans. Report of the joint working group on lane rental for local authority roads. (19R9). Assoc. of County Councils, U.K. Srinivasan, R., and Harris, F. C. (1991). "Lane rental contracting." J. Constr. Mgmt. and Economics, 9(2), Reading, U.K., 191-195. Tarricone, P. (1993). "Deliverance." Civ. Engrg., ASCE, 63(2), 36-39.
JOURNAL OF CONSTRUCTION ENGINEERING AND MANAGEMENT / SEPTEMBER 1995/281
J. Constr. Eng. Manage., 1995, 121(3): 273-281
