Development of life cycle costing framework for highway ... - CiteSeerX

INTERNATIONAL JOURNAL OF

PROJECT MANAGEMENT International Journal of Project Management 23 (2005) 37–44 www.elsevier.com/locate/ijproman

Development of life cycle costing framework for highway bridges in Myanmar D. Singh, Robert L.K. Tiong

*

School of Civil and Enviromental Engineering, Centre for Advanced Construction Studies, Nanyang Technological University, Nanyang Avenue, Singapore 639798 Received 12 December 2003; received in revised form 2 April 2004; accepted 18 May 2004

Abstract The aim of any engineering design is to minimize the total cost of the structure without compromising the functional requirements while maximizing the utility of the structure to the users in particular and to the society in general. Life cycle costing is a technique for determining the most effective capital investment option for achieving technical-economic optimization of a structure/ system. This paper briefly describes a detailed procedure for developing a framework for life cycle costing analysis (LCCA) of highway bridges in Myanmar. The paper discusses various cost components and other statistical factors that need to be taken into consideration while assessing the life cycle cost (LCC) of a highway structure. A stepwise procedure to determine various cost components that come into LCC calculation is also illustrated. The effect of uncertainties associated with various factors on the total cost of the structure is demonstrated performing sensitivity analysis. An attempt is also made to demonstrate how better quality construction with increased initial cost can lead to lower LCC of a highway structure. The study has made a call for the development of comprehensive life cycle costing framework for transportation-related projects in Myanmar in order to be able to strike a balance between the need for maintenance and replacement of highway structures and limited funds available for their upkeep. Ó 2004 Elsevier Ltd and IPMA. All rights reserved. Keywords: Life cycle cost; Cost optimization; User cost; Highway structures

1. Introduction A nation’s economic strength is reflected in its infrastructure assets. The capability of any developing country to produce and sustain economic growth is directly related to its ability to transport the goods and services that it produces. A good road network/system is very important to economic activities of a nation as it plays a pivotal role in disbursing basic public services like food distribution, water supply, waste removal, and medical facilities both efficiently and economically. The success of such a system depends on the ability of policy makers to strike a balance between available resources and the need for creation of new facilities and maintenance and repair of existing infrastructure components. *

Corresponding author. Tel.: +65-679-04913; fax: +65-679-16697. E-mail address: [email protected] (R.L.K. Tiong).

0263-7863/$30.00 Ó 2004 Elsevier Ltd and IPMA. All rights reserved. doi:10.1016/j.ijproman.2004.05.010

Recent developments in Myanmar have led to a rapid, continuous increase in traffic volumes on both urban and rural roads. The need for renovating old bridges and constructing new ones has consequently increased. According to Public Works Department (PWD) year 2001 database, Myanmar has a road network with more than 34,600 bridges (including culverts) totaling a length of approximately 550 km. The nation still needs an extensive network of highways and bridges to make its infrastructure system more efficient and the Government has been giving special focus on the development of the national road and bridge network to maintain its present growth rate of economy [1]. As these structures approach their designed service life, a plan has to be implemented to repair or rehabilitate them. These repair and rehabilitation operations consume significant part of the limited funds available. The efficient use of these available funds can be achieved by

38

D. Singh, R.L.K. Tiong / International Journal of Project Management 23 (2005) 37–44

optimizing the whole life cost of structures rather than just initial cost of construction only. Therefore, in order to strike a balance between available funds and the need for repair and replacement of these infrastructure components, it is important to develop a comprehensive technical-economic optimization methodology that takes into account lifetime management (inspection, repair, rehabilitation and replacement) costs in addition to initial cost of construction. The aim of any engineering design is to minimize the total cost of the structure without compromising the functional requirements while maximizing the utility of the structure to its users in particular and to the society in general. The client and designer together play an important role in making the cost effective and environmentally conscious choices in the selection of the design and construction method and materials, which significantly influences both resource consumption and management perspectives, substantially impacting annual operation and maintenance costs. Therefore, infrastructure design should be viewed from life cycle costing perceptive as it has the greatest potential of minimizing a structure’s whole life cost during planning and design stage at which changes are most easily implemented and where resistance to change is the least [2]. Life cycle costing is an economic assessment of an item, area, system or facility considering all costs of ownership over an economic life, expressed in terms of equivalent dollars [3]. It takes into account time value of money and reduces a flow of running costs over a period of time to a single current value or present worth (PW). Life cycle costing can be used as a management tool or as a management system [4]. As a management tool it can be used intermittently throughout the economic life of the structure, whenever different options are available, to determine the alternative with the lowest LCC. On the other hand, as a management system in continuous operation it can be used to actively manage the asset throughout its service life. LCCs estimated at one stage are carried through into the budget for the next stage. Operation and maintenance costs constitute a major portion of the total LCC of a structure [5]. One way to create a more comprehensive view of costs in different phases of a civil engineering project is to perform LCCA. LCCA takes into account all aspects of lifetime cost of the system such as agency costs and the impacts of the system on the users in particular and on the society in general. The main motivation to use LCCA is to increase the possibility of cost reductions during operation and maintenance even if that means spending somewhat more during planning and development [2]. Comprehensive structural optimization requires a lifetime perceptive, that is, the explicit consideration of design, construction, services, inspection, maintenance and decommissioning [6]. Several researchers have de-

veloped methodologies for bridge design and management system based on lifetime costs and benefits (e.g. [7– 9]). Bridges and roads are important components of infrastructure system without which basic public services cannot be efficiently disbursed. Repairs and maintenance of these components consume a lot of resources. Many researchers and practitioners, for e.g., [10,11], have proposed optimal maintenance strategies for critical bridge elements. Frangopol et al. [12] and Yanev [13] expressed the importance of the need for the application of LCCA method to maintenance planning decisions of the bridges, which are on the brink in the US national highway system. The purpose of this paper is to demonstrate a stepwise procedure for developing a comprehensive framework for LCCA of highway bridges in Myanmar. This paper also aims at encouraging the application of life cycle costing approach to transportation-related projects in Myanmar so that the optimal situation is arrived at more often than not.

2. Methodology A highway reinforced concrete (RC) bridge, newly constructed using suspended cantilever method, is selected for the study. The reason for selecting this bridge for the study is that, in Myanmar most RC bridges are being constructed using this method. LCC components of highway structures comprise agency costs, user costs, accidents and external cost components. Even though this study attempts to integrate initial cost of construction and other components of LCC in predicting the total cost associated with a bridge over its intended service life, accident costs and external cost components are not included in the LCCA of the Example Bridge due to insufficient statistical data. A good illustration of estimation of these costs can be followed in [14,15]. The authors collected all cost data at different levels of detail from bills of quantities (BoQ) of the Example Bridge, maintenance season average daily traffic (MSADT) volumes on the highway and maintenance history of some of RC bridges in the country. The authors emphasize on the factors such as delay time cost, additional fuel consumption and maintenance costs incurred whenever a vehicle is caught in congestion, as shown in Eq. (1): Road User Cost ¼ CDT þ CAF þ CAM;

ð1Þ

where CDT represents cost of delay time, CAF represents cost of additional fuel consumption and CAM represents cost of additional maintenance of the vehicle. The impacts on LCC of uncertainties associated with input parameters such as discount rate, MSADT and work zone duration are also studied by performing sensitivity analysis. All cost data are normalized to the equivalent US dollars in the year 1999.


3. LCCA of the example bridge In performing the LCCA of highway projects, all aspects of LCC of the system such as agency costs, performance of the facility, maintenance and rehabilitation, social and economic impacts of the system on the users should be taken into consideration. 3.1. Agency/owner cost components This component includes costs incurred by the agency or owner over the lifetime of the facility. They include initial cost, rehabilitation cost, prevention or routine maintenance cost and cost of disposal of the structure at the end of service life. 3.2. Road user costs Road user costs are those costs incurred by the users of the system due to maintenance operation of highway structure causing congestion or disruption of normal traffic flow. These costs are not directly measurable but the traffic delays that lead to these costs can be measured. Traffic delay costs have, therefore, to be predicted on the basis of estimated delays and vehicle operation cost which includes additional cost of fuel plus additional cost of maintenance. 3.3. Delay time cost Travel time delay normally makes a significant contribution to the user costs component, since the value of unit time and the duration of additional time spent in congestion are multiplied to estimate the total cost of commuters travelling on the road. To calculate delay time cost, the occupancy value (OCV) in each type of vehicle in Myanmar is required to estimate the total number of commuters on the road. From a spokesperson 1 of the local bus services, the average occupancy value for buses on a normal non-peak period is about 60%. The spokesperson 2 from local taxi industry estimated that the average number of people in a taxi is about 3. The authors assumed that different professions are expected to travel in different types of vehicles. For example, commuters in the buses could come from all walks of life. For those travelling in taxis, they could be either professionals or holding equivalent positions that command a salary of approximately $600 per month. Using Eq. (2), average salary of commuters travelling in each type of vehicles can be calculated.

P AvSali ¼

pij Ej MGWj P ; pij Ej

39

ð2Þ

where AvSali represents the average monthly salary of commuters travelling in vehicle type i, and pij represents the proportion of commuters with profession j in vehicle type i, Ej represents the total employment in profession j and MGWj represents the mean gross wage of profession j. Table 1 shows the authors’ assumption, on the basis of the finding of the survey of 2000 commuters travelling in different types of vehicles by the first author of the paper, of the proportion of each profession travelling in different types of vehicles and their average salary. Now, the cost of delay time per minute can be calculated as shown in Table 2. Average salary per month for each commuter travelling in different types of vehicles is normalized to average salary per minute by dividing the values with 22 working days per month, eight working hours per day and 60 min/h. Then, value of time lost or cost of delay time is obtained by multiplying the average salary per minute with occupancy rate for different types of vehicles respectively. 3.4. Vehicle operating cost These costs are incurred by the vehicle owner due to the blockage caused by maintenance operation of highway structures. The disruption of normal traffic flow causes reduced speed, frequent changing between acceleration and deceleration actions and queuing that affect the consumption of fuel and oil, tire wear, vehicle maintenance, vehicle depreciation, and spare parts. 3.5. Additional cost of fuel In congestion, the consumption of fuel is much higher because of the start-and-stop effect on vehicles. Fuel consumption of vehicles in a smooth traffic flow is lesser than those in an irregular flow. The more interchanging between the acceleration and braking actions occurs, the more fuel consumption increases. To calculate the cost of additional consumption of fuel for all types of vehicles, factors such as the cost per liter of fuel, the estimated efficiency factor of fuel consumption and the average distance covered by a liter of fuel are required. Local automobile experts 3 have estimated that the usage of fuel is six to ten times higher for a vehicle travelling in an irregular traffic flow or slow-moving traffic as compared to the vehicle advancing at consistent speed. Therefore, an average value of 8 is assumed for the calculation of additional fuel consumption and maintenance cost resulting from the congestion of traffic

1

The spokesperson is traffic inspector of the local bus industry. The spokesperson is marketing executive of the local taxi services company. 2

3

The marketing executives of local Suzuki automobile company.

40


Table 1 Estimated proportion of different professions travelling in different types of vehicles Vehicle type, i (MGW)

1. 2. 3. 4. 5.

Types of professions, j

Private cars ($722.92) Taxis ($631.13) Buses ($327.53) LGV ($206.76) HGV ($206.76)

1

2

3

4

5

6

7

8

0.6 0.3 0.1 0.0 0.0

0.6 0.3 0.1 0.0 0.0

0.3 0.4 0.3 0.0 0.0

0.0 0.4 0.6 0.0 0.0

0.0 0.3 0.7 0.0 0.0

0.0 0.3 0.7 0.0 0.0

0.0 0.1 0.3 0.3 0.3

0.0 0.1 0.3 0.3 0.3

LGV ¼ Light goods vehicle; HGV ¼ Heavy goods vehicle; 1 ¼ Managers; 2 ¼ Professionals; 3 ¼ Technicians; 4 ¼ Craftsmen; 5 ¼ Clerks; 6 ¼ Sale & services workers; 7 ¼ Operators; 8 ¼ Laborers.

Table 2 Value of time lost for different types of vehiclesa

Table 3 Maintenance cost for different types of vehicles

Vehicle type

OCV, a

AvSal ($)

AvSal/min, b ($/min)

Value of time lost a b ($/min)

Vehicle types

Maintenance cost ($/year)

Private cars Taxis Buses LGV HGV

1.51

722.92

0.07

0.10

631.13 327.53 206.76 206.76

0.06 0.03 0.02 0.02

0.18 1.67 0.03 0.03


913.25 992.50 2203.50 2015.50 2080.75

3.0 54 1.4 1.4

a Occupancy rates for cars and goods vehicles are quoted from Natsinas and Mintsis [16].

flow due to work zone implemented for maintenance operation of the structure. The cost of travel per kilometer (km) can be calculated from the distance travelled with a liter of fuel, cost of fuel per liter and fuel efficiency. For example, the cost of gasoline in 1999 was about $1.2 per liter and the average distance travelled by a car is 15 km/l 3 the cost of travel per kilometer for a car can be calculated as [$1.2/l)/(15 km/l) 8] as $0.64 per km. Similarly, the cost of travel per km can be calculated as $0.64 for taxis and $1.60 for buses and goods vehicles. 3.6. Additional cost of maintenance Additional cost of maintenance incurred by a vehicle caught in a congestion is due to the higher wear and tear of car components such as brake pads, clutch plates, etc. Certain components of vehicles, which are directly related to the type of travelling conditions on the roads, are selected for the study purpose. These cost items include cost of servicing, tires, brake pads, clutch plates, timing belt, air-conditioner, exhaust, alternator and batteries. These items need faster servicing or replacement when vehicles have to travel through the congested traffic more frequently. The frequent braking and accelerating would result in higher wear and tear of tires, clutches and brake pads. When vehicles are trapped in the traffic queue, the higher temperature also affects the alternator and batteries. A vehicle requires regular servicing after a certain mileage. Regular servicing does not include servicing of those components that are worn out

Mileage (km/year) 12,000 15,000 109,000 180,000 187,000

Maintenance cost ($/km) 0.08 0.08 0.02 0.01 0.01

due to usage. For example, in the case of a car, the tires get worn out after travelling for an average distance of 15,000 km. From the data obtained from a local automobile service station, 4 the average mileage a car travels in a year is about 12,000 km. Hence, the tires need to be replaced once in every 1.25-year time interval. As the cost of each tire was about $45 in 1999, the cost of four tires would be around $180. Therefore, the maintenance cost per year for tires would amount to $144. Column 2 of Table 3 shows the maintenance cost for different types of vehicles. Now, cost of maintenance per kilometer of travel for different types of vehicles is obtained by dividing total cost of maintenance per year with average annual mileage. They are summarized in Table 3. Assuming that all types of vehicles cruise at an average speed of 55 km/h on the highway, the distance traversed by each vehicle in 1 min would be approximately 0.917 km. Table 4 shows vehicle operating cost (VOC) per minute for different types of vehicles. From the table, it is clear that VOC per minute for cars and taxis are almost same and those for buses and goods vehicles do not differ significantly. Summing up, user delay cost for different types of vehicles can be worked out as illustrated in Table 5. It is clear from the Table 6 that taxis and buses together represents only about 4% of the maintenance season average daily traffic volume on the highway. If types of vehicles were to be categorized as ‘car’ and ‘truck’ only, category ‘car’ would represent all private

4

It is a local Suzuki automobile service station.

D. Singh, R.L.K. Tiong / International Journal of Project Management 23 (2005) 37–44 Table 4 Operating cost for different types of vehicles Vehicle types

Cost of travel, a ($/km)

Maintenance cost, b ($/km)

VOC, (a þ bÞ 0:917 ($/min)


0.64

0.08

0.66

0.64 1.60 1.60 1.60

0.08 0.02 0.01 0.01

0.65 1.49 1.48 1.48

41

mar is little more than 50 years in practice, the authors considered an analysis period of 40 years only owing to the fact that the power of discounting is such that the present value of $1 in 40 years’ time, using a discount rate of 8% 5 is only about 4.6 cents, so that any costs incurred after 40 years can normally be considered to be insignificant. For calculation of LCC, the following equation is used. PW of LCC ¼ Initial Cost þ PW ðMaintenance Cost þ User CostÞ:

Table 5 User cost for different types of vehicles Vehicle types

Delay time cost, a ($/min)

VOC, b ($/min)

User cost, (a þ b) ($/min)


0.10

0.66

0.76

0.18 1.67 0.03 0.03

0.65 1.49 1.48 1.48

0.83 3.16 1.51 1.51

Table 6 MSADT volume of different types of vehicles Vehicle type Private cars Taxis Buses LGV HGV MSADT volume

1998 6382 177 216 1827 1913 10,515

1999 6584 198 233 1626 1952 10,593

2000 7426 207 240 1763 1963 11,599

cars and taxis and category ‘truck’ would represent all the other types of vehicles and constitute approximately 35% of MSADT. Now, the user cost for category ‘car’ can safely be taken as $0.76/min and for ‘truck’ category as $1.5/min. 3.7. Calculation of LCC of the Example Bridge The total deck area of the bridge is approximately 3345 square-meter. Based on the PWD database, it can be assumed that RC bridges in the local road network generally require the first major repair operation at about 20-year age and subsequent rehabilitation at about 35-year age; duration of repair operation generally varies from 6 to 12 weeks depending upon the extent of the repair work needed; and the budget, inclusive of traffic management cost, allotted for the routine maintenance work is approximately $25 and for major repair work approximately $100 per square-meter of bridge deck. Though the average life of RC bridges in Myan-

ð3Þ

In the calculation of LCC of the bridge, it is essential that the risks and uncertainties associated with statistical parameters such that discount rate, traffic flow, defect occurring frequency and duration of maintenance operation be properly considered. In order to analyze the sensitivity of LCC with changes in discount rate (R) and traffic volume – the two statistical factors that LCC is considered to be more sensitive to, total LCC of the bridge is calculated for varying traffic volumes using different discount rates. The sensitivity of duration of work zone for maintenance operation is also included by considering 8- and 12-week repair works period at 8% discount rate and varying traffic flows. 3.8. Sensitivity analysis of data Life cycle cost analysis can be greatly enhanced by using other supplemental analysis techniques to gain a greater understanding of impacts of LCC parameters on the total cost of the structure. Sensitivity analysis is a technique for evaluating how stability of the result or outcome depends on the variation in various input parameters. Fig. 1 shows the variation in total LCC of the Example Bridge for varying traffic flows at different discount rates. The graph clearly indicates that at lower discount rate, the proportion of the road user cost components in total LCC of the bridge for heavy traffic flow is quite significant. At lower volume of MSADT about 10,000 or less sensitivity of total LCC to discount rate in not much significant. Fig. 2 illustrates the effect of duration of maintenance operation on the LCC of the Example Bridge. In the graph, LCCs of the bridge for 8- and 12-week works duration of maintenance operation are shown. From the graph, it is clear that at lower volume of traffic, the effect of work zone duration is not significant. But, when MSADT is about 120,000 LCC of the bridge is increased by more than 20%, which means that at higher level of traffic flows duration of repair period also plays significant role in analyzing different alternatives of bridge 5 PWD currently uses an interest rate of 8% for economic assessment of its projects.

42


LCC(PW) ($) 8,000,000 7,000,000 6,000,000 5,000,000 4,000,000 3,000,000 2,000,000 1,000,000 0 10000

20000

R=4% R=8%

40000 R=6% R=10%

80000

120000

MSADT(Number)

Fig. 1. PW of total LCC of the Example Bridge for varying MSADT volumes at different discount rates.

ered have more or less the same expected service life, the uncertainty associated with the forecast would be common to all alternatives and the effect of variation in the expected service life will not make any difference to the total LCC of the bridge. For comparing alternate bridge designs having different analysis period or service life, it is important that total LCC be converted into equivalent uniform annual cost (EUAC) instead of present worth and the alternative that gives the lowest EUAC should be considered as the optimal option. As pointed out by Stone [17] ‘‘Errors of 5 or 10 years in the predicted life will not make very much difference to the predicted equivalent costs when the life is 50–60 years or more. The shorter the life, the greater the effect of error in determining it. The errors in predicted costs, and hence in design decisions, are likely to be greater when the life of the building is taken as substantially shorter than conditions warrant than when the life is taken as longer than justified.’’ Most RC bridges in Myanmar have a normal expectation of service life of about 50 years. 3.9. Example to appraise alternate options

LCC(PW) ($)

6,000,000 5,000,000 4,000,000 3,000,000 2,000,000 1,000,000 0 10000 8-week works

20000

40000

12-week works

80000

120000

MSADT (Number)

Fig. 2. PW of total LCC of the Example Bridge for 8-week and 12week maintenance works at varying traffic flows.

design and maintenance program. This highlights the fact that sensitivity with respect to one parameter can be affected by the value of another parameter. It is also clear from the example that the lower the discount rate the more sensitive is LCC to the traffic flow. Similarly, the higher the volume of MSADT the more sensitive is LCC to the work zone duration. The parameter ‘economic life’ has a great influence on the LCC of a facility owing to the exponential nature of its effect in LCC calculation. If the alternatives consid-

The original design of the bridge uses the Grade 30 Portland cement concrete (PCC). The research done by Righden et al. [18] revealed that the use of high-grade cement concrete is likely to reduce the degeneration rate by 40%. No such models capable of predicting the deterioration frequency have been developed for the bridge stock in Myanmar. So, based on that research the authors assumed that due to high durability of Grade 45 PCC used for bridge deck the initial defect free period of the bridge would be extended from 20 to 25 years. Based on the information in Table 6, the MSADT on the highway is estimated to be around 40,000 after 25 years when work zone would be implemented for rehabilitation operation. Initial cost of the bridge is obtained by calculating the cost of the bridge deck using PWD’s rate for Grade 45 PCC in place of Grade 30 PCC and PW of LCC is obtained using Eq. (3). Table 7 shows the cost savings resulting from the use of high-grade cement concrete. This illustration is, in essence, an attempt by the authors to demonstrate the application of life cycle costing approach to the decision-making during design stage, and to highlight how the use of better quality construction with increased initial cost could lead to lower LCC of highway bridges in Myanmar. A call is, thereTable 7 Cost summary of alternative design of bridge deck Cost summary

PCC Grade 30

PCC Grade 45

Savings ($)

Initial cost Other LCC (PW)

$342,631 $958,468

$352,847 $574,914

)$10,216 $383,554

Total cost savings

$373,338


fore, made to develop a reliable database system to predict the degeneration rate of bridges in the national highway system and to develop a highway maintenance manual like The Rural Road Maintenance Manual developed by Department for Transport, the UK.

4. Conclusion and recommendation The study discusses the various cost components that come into the calculation of LCC of a bridge and proposes a framework for LCC analysis of highway structures in Myanmar. As the road user costs associated with highway bridge construction usually exceeds, particularly in urban areas where traffic flow is high, the initial cost of construction by a substantial amount, minimization of the disruption of traffic flow during each maintenance operation throughout the service life is an important aspect of any highway bridge design. The study highlights the statistical parameter to which LCC of a bridge is more sensitive. At lower traffic volume, the effect of user delay cost is not significant. However, at lower discount rate and heavy traffic flow the initial cost of construction is in essence irrelevant. The relevance of the result of a LCC calculation is often considered to be somewhat uncertain. This is mainly due to lack of sufficient cost data and accepted industry standards for describing the life cycle behaviors of facility and internal processing system [19]. Therefore, the challenge in using this framework is quantifying uncertainties in the input variables such as discount rate and defect occurrence frequency, which greatly influence the total cost of the structure. As the outcome of any LCC exercise is only as good as the assumptions upon which these predictions are based, effective life cycle costing needs an equally effective risk management system in order to use risks and uncertainties to improve decision-making. It is, therefore, vitally important to develop a reliable database system on which more precise estimation of these parameters can be based. One major shortcoming of the proposed methodology is that it does not cover all the costs that are incurred by the project. The inclusion of the accident cost component and the external cost component may have great impact on the total cost of the structure and influence on the final selection of the design options and maintenance programs. However, this paper has advanced an awareness of the ability of LCCA to capture these costs, with an expectation that other costs may be identified and predicted in the future. Such costs may include the effects of excess emission, accidents, increased noise, or heightened driver stress that can lead to more accidents. The study has also made a call for the development of a model capable of predicting the defect occurrence frequency of bridge components and a traffic model capable of comprehensively defining as to how formation

43

of queue of traffic due to congestion could influence the selection of alternate bridge designs in the local context so that the proposed framework can be further developed into a comprehensive value analysis framework for a wide range of transportation-related projects. The approach presented in this paper is somewhat crude and imprecise. If viewed from proper perceptive, i.e., a decision support tool to be considered with engineering judgments and other factors, even a crude LCCA based on educated professional guesses will usually lead to better decisions than no LCC considerations at all. As noted by Bondstedt [20], LCCA can be, in any particular instance, ‘‘specifically wrong’’, but overall ‘‘generally right’’. This makes LCCA an essential tool to ensure efficient use of limited funds available for the infrastructure development of any developing nation like Myanmar. Therefore, policies and practices need to be developed to promote the integration of maintenance, repair and replacement decisions in the design of highway structures.

References [1] http://www.unescap.org 12 March 2001. [2] Kirk S, Dell’Isola AJ. Life cycle costing for design professionals. second ed. NY, USA: McGraw-Hill; 1995. [3] Dell’Isola AJ. Value engineering in the construction industry. 3rd ed. USA: Van Nostrand Reinhold; 1982. [4] Ferry DJO, Flanagan R. Life cycle costing – a radical approach. Construction Industry Research and Information Association, Report 122, 1991. [5] Flanagan R, Norman G. Life cycle costing: Theory and practices. London, UK: RICS Surveyors Publication Ltd.; 1989. [6] Jiang M, Corotis RB, Ellis JH. Optimal life cycle costing with partial observability. J Infrastruct Systems 2000;6(2):56–66. [7] Tao Z, Corotis RB, Ellis JH. Reliability-based structural design with Markov decision processes. J Struct Eng, ASCE 1995;121(6):971–80. [8] Hearn G, Shim H-S. Integration of bridge management systems and nondestructive evaluations. J Infrastruct Systems, ASCE 1998;4(2):49–55. [9] Esters AC, Frangopol DM, Lin K-Y. Minimum expected life cycle cost design for bridges. In: Proceedings of the Seventh IFIP WG75 Working Conference, Reliability and Optimization of Structural Systems. Terrytown, NY: Pergamon; 1997. p. 133–40. [10] Vorster MC, Bafna T, Weyers RE. Model for determining the optimum rehabilitation cycle for concrete bridge decks. Transportation and Research Record 1319, Transportation Research Board, National Research Council, Washington D.C., 1991. p. 62–7. [11] Markow MJ, Madanat SM, Gurenich DI. Optimal rehabilitation times for concrete bridge decks. Transportation and Research Record 1392, Transportation Research Board, National Research Council, Washington D.C., 1993. p. 69–70. [12] Frangopol DM, Ghosn M, Hearn G, Nowak A. Guest editorial: Structural reliability in bridge engineering. J Bridge Eng, ASCE 1998;3(4):151–4. [13] Yanev B. The management of bridges in New York city. Eng Struct 1998;20(11):1020–6. [14] Andreassend DC. A framework for accident costing. Aust Road Res Board 1983;13(4):300–1.

44


[15] Matthews HS, Hendrickson C, Horvath A. External costs of air emissions from transportation. J Infrastruct Systems 2001; 7(1):13–7. [16] Natsinas T, Mintsis G. Important traffic characteristics in Thessalonica, Greece: A preliminary study. Traffic Eng Contr 1994;11. [17] Stone PA. Building design evaluation: Cost-in-use. Cambridge: The University Printing House; 1980. p. 59.

[18] Rigden SR, Burley E, Tajali SMA. Life cycle costing and the design of structures with particular references to bridges. Proc Inst Civil Eng: Municipal Engineer 1995;109:284–8. [19] Abraham D, Dickinson R. Optimal costs for environmentally regulated facilities. J Construct Eng Manage 1998;124(2):146–54. [20] Bondstedt HO. Specifically wrong – but generally right! Actual life expectancy projections for bridges. Pre-stressed Concrete Association of Pennsylvania, May 1997.