Optimal maintenance strategies for bituminous pavements: A case study in Vietnam using HDM-4 with gradient methods

 advisable before the pavement reaches a very poor condition (IRI ³ 
8). The results also showed that DBST did not improve the roughness condition, but it 
is effective in delaying the deterioration progress, as a result of obtaining the positive 
NB value in comparison with the base case (only routine maintenance). 
6) As indicated in the last column of Table 7, the net benefit gained over a 20-year 
2 Re-DBST is also suggested for pavements with lower initial roughness with low traffic but it was 
applied at later years when roughness was predicted to reach 8 IRI. 
Journal of the Eastern Asia Society for Transportation Studies, Vol. 6, pp. 1123 - 1136, 2005
1131
analysis period was much larger for high and medium traffic roads than for low traffic 
roads, especially for the roads in poor and very poor condition, because for the low 
traffic roads with the absence of large buses and heavy trucks (as commercial basis for 
operating cost), the vehicle operating cost savings as a result of road improvements are 
much smaller for two-wheelers and passenger cars, and so leading to low net benefit 
gained over the analysis period. This finding suggests that the priority should be given 
to the roads with high or medium traffic levels, if funds are not available to provide 
optimum maintenance to all the traffic levels. 
Table 7. Optimum Maintenance Options 
High Traffic (AADT = 10000vpd) & HRA 
Initial 
IRI 
k 
(1) 
Optimum Maintenance Options 
(2) 
NB ($US mil) 
(3) 
2 2 [2015, 2021; Ovl-30, Ovl-30] 25.08 
3 2 [2008, 2016; Ovl-30, Ovl-30] 28.35 
4 3 [2004, 2011, 2018; Ovl-30, Ovl-30, Ovl-30] 51.92 
5 3 [2004, 2013, 2019; Ovl-50, Ovl-30, Ovl-30] 60.69 
6 3 [2004, 2013, 2019; Ovl-60, Ovl-30, Ovl-30] 70.83 
7 3 [2004, 2014, 2020; Ovl-70, Ovl-30, Ovl-30] 82.38 
8 3 [2004, 2013, 2019; Ovl-70, Ovl-30, Ovl-30] 95.28 
9 3 [2004, 2014, 2021; Re-50, Ovl-30, Ovl-30] 108.47 
10 3 [2004, 2014, 2021; Re-50, Ovl-30, Ovl-30] 123.08 
Medium Traffic (AADT = 4000vpd) & SBSD 
 (1) (2) (3) 
2 1 [2012; Ovl-30] 25.53 
3 1 [2012; Ovl-30] 26.05 
4 2 [2004, 2016; Ovl-30, Ovl-30] 34.96 
5 2 [2004, 2014; Ovl-40, Ovl-30] 50.63 
6 2 [2004, 2014; Ovl-50, Ovl-30] 53.67 
7 2 [2004, 2012; Re-DBST, Ovl-30] 55.95 
8 2 [2004, 2012; Re-DBST, Ovl-30] 59.48 
9 2 [2004, 2012; Re-DBST, Ovl-30] 63.25 
10 2 [2004, 2012; Re-DBST, Ovl-30] 65.90 
Low Traffic (AADT = 1000vpd) & SBSD 
 (1) (2) (3) 
2 1 [2016; DBST] 0.62 
3 1 [2016; DBST] 0.65 
4 2 [2010, 2020; DBST, DBST] 5.71 
5 1 [2009; Re-DBST] 6.02 
6 1 [2009; Re-DBST] 6.19 
7 1 [2007; Re-DBST] 6.75 
8 1 [2004; Re-DBST] 7.11 
9 1 [2004; Re-DBST] 7.70 
10 1 [2004; Re-DBST] 8.39 
Note: (1) “k” is the number of maintenance works applied over the analysis period 
(2) Ovl-30, Ovl-40, Ovl-50, Ovl-60, and Ovl-70 denote overlays with thicknesses of 30, 40, 50, 
60, and 70 mm, respectively; Re-DBST and Re-50 denote reconstruction with DBST of 25 mm 
thickness and with 50 mm- asphalt concrete, respectively. 
Journal of the Eastern Asia Society for Transportation Studies, Vol. 6, pp. 1123 - 1136, 2005
1132
(3) NB is the maximum net benefit over the analysis period of the optimal maintenance option, 
where net benefit being defined as the saving in total transport cost over the base case. 
Table 8. Suitable Maintenance Works To Be Applied in the First Year 
for Pavements with Different Traffic Levels & Initial Conditions 
Initial Pav. 
Condition 
Traffic Level & Existing Pavement Type 
Condi- 
tion 
IRI Low Traffic & SBSD 
(AADT = 1000 vpd) 
Medium Traffic & SBSD 
(AADT = 4000 vpd) 
High Traffic & HRA 
(AADT = 10000 vpd) 
2 Good 
3 
Routine Maintenance 
4 Overlay 30 mm Fair 
5 Overlay 40 mm Overlay 50 mm 
6 Overlay 50 mm Overlay 60 mm Poor 
7 
8 
Overlay 70 mm 
9 Reconstruction with DBST 
Very 
poor 
10 
Reconstruction with 
50 mm-asphalt concrete 
Figure 2. Predicted Roughness Progression under Optimal Maintenance Options 
High Traffic Case (AADT = 10 000 vpd) 
1
2
3
4
5
6
7
8
9
10
11
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
20
17
20
18
20
19
20
20
20
21
20
22
20
23
Time (year)
 I
nt
er
na
tio
na
l R
ou
gh
ne
ss
 In
de
x 
(IR
I) 
 .
Initial IRI = 2 Initial IRI = 3 Initial IRI = 4
Initial IRI = 5 Initial IRI = 6 Initial IRI = 7
Initial IRI = 8 Initial IRI = 9 Initial IRI = 10
Journal of the Eastern Asia Society for Transportation Studies, Vol. 6, pp. 1123 - 1136, 2005
1133
Figure 4. Predicted Roughness Progression under Optimal Maintenance Options 
Low Traffic Case (AADT = 1000 vpd) 
1
2
3
4
5
6
7
8
9
10
11
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
20
17
20
18
20
19
20
20
20
21
20
22
20
23
Time (year)
In
te
rn
at
in
al
 R
ou
gh
ne
ss
 In
de
x 
(IR
I) 
 .
Initial IRI = 2 Initial IRI = 3 Initial IRI = 4
Initial IRI = 5 Initial IRI = 6 Initial IRI = 7
Initial IRI = 8 Initial IRI = 9 Initial IRI = 10
Figure 3. Predicted Roughness Progression under Optimal Maintenance Options 
Medium Traffic Case (AADT = 4000 vpd) 
vpd) 
1
2
3
4
5
6
7
8
9
10
11
20
04
20
05
20
06
20
07
20
08
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
20
17
20
18
20
19
20
20
20
21
20
22
20
23
Time (year)
In
te
rn
at
io
na
l R
ou
gh
ne
ss
 In
de
x 
(IR
I) 
 .
Initial IRI = 2 Initial IRI = 3 Initial IRI = 4
Initial IRI = 5 Initial IRI = 6 Initial IRI = 7
Initial IRI = 8 Initial IRI = 9 Initial IRI = 10
Journal of the Eastern Asia Society for Transportation Studies, Vol. 6, pp. 1123 - 1136, 2005
1134
5. CONCLUSIONS 
The paper has presented a systematic approach to investigate the suitable maintenance 
strategy for the pavement management system by using HDM-4 with gradient methods 
corresponding to the traffic levels and initial pavement conditions of bituminous pavements in 
the flat terrain of Vietnam. From this study, it was found that reconstruction was much more 
effective for badly deteriorated pavements, while other maintenance works were advisable 
depending on the levels of traffic and initial pavement conditions. Also, thin overlays of 30 
mm along with routine maintenance were much more effective to maintain the pavements of 
high and medium traffic roads in good condition (IRI £ 3), while the DBST along with routine 
maintenance were found to be the optimum for the pavements of low traffic roads in good or 
even in relatively poor condition (IRI £ 7). In addition, the findings showed that the priority 
should be given to the roads with high or medium traffic levels, if funds are not available to 
provide optimum maintenance to all the traffic levels. The investigated results, thus, may help 
road agencies in better selecting the right maintenance works based on condition surveys and 
to pay more attention to routine maintenance to preserve the road network in appropriate ways. 
If detailed data are available and all sensitive parameters are properly calibrated according to 
the specific conditions of the considered regions, further examinations should be carried out 
to obtain more appropriate results, including different types of pavements and terrain 
conditions. 
REFERENCES 
AASHTO. (1993) Guide for Design of Pavement Structures. American Association of State 
Highway and Transportation Officials. Washington D.C. 20001. 
Bennett, C.R. and W.D.O. Paterson (2002) A Guide to Calibration and Adaptation. Volume 
V, HDM-4 Manual. World Road Association, ISOHDM, PIARC, Paris, France. 
HMCP (2002). Highway Management Capacity Improvement Project. The Final Report, 
Vietnam Road Administration, Ministry of Transportation of the Socialist Republic of 
Vietnam. 
HDM-4. (2003) Highway Development and Management Model, Version 1.3. World Road 
Association, ISOHDM, PIARC, Paris, France. 
RNIP. (2003) 10-Year Strategic Plan for National Road Network, Road Network 
Improvement Project. The Report, Vietnam Road Administration, and Ministry of 
Transportation of the Socialist Republic of Vietnam. 
Rolt, J. and C.C. Parkman (1997) Characterisation of Pavement Strength in HDM-III and 
Changes Adopted for HDM-4. TRL Project Report PR/ORC/587/97, Transport Research 
Laboratory, Crowthorne, UK. 
RTP2 (2002). Second Rural Transport Project. The Field Study Report. Ministry of 
Transportation of the Socialist Republic of Vietnam. 
TCVN 4054. (1998) The Highway Design Standards. Ministry of Transportation of the 
Socialist Republic of Vietnam. 
Journal of the Eastern Asia Society for Transportation Studies, Vol. 6, pp. 1123 - 1136, 2005
1135
The World Bank. (2003) Project Appraisal Document for Road Network Improvement 
Project in Vietnam. Report No. 27368-VN, The World Bank, Washington, D.C. 
Tsunokawa, K. and R. Ul-Islam (2003). Optimal Pavement Design and Maintenance Strategy 
for Developing Countries: An Analysis Using HDM-4. International Journal of Pavement 
Engineering, Vol. 4, pp.193-208. 
Tsunokawa K., D.V. Hiep, and R. Ul-Islam (2005). True Optimization of Pavement 
Maintenance Options Using What-If Models: An Application of Gradient Search Methods 
with HDM-4. Proc. of the 84th Transportation Research Board Conference, Washington, 
D.C. 
Ul-Islam, R. and K. Tsunokawa (2004). Identification of Appropriate Routine Maintenance 
Works for Main Roads and Local Roads using HDM-4. Proc. of the 6th International 
Conference on Managing Pavements (ICMP-6), Brisbane, Queensland, Australia. 
Watanatada T., C.G. Harral, W.D.O. Paterson, A.M. Dhareshwar, A. Bhandari, and K. 
Tsunokawa (1987) The Highway Design and Maintenance Standards Model, Volume 1: 
Description of the HDM-III Model. World Bank Publications, Washington, D.C. 
Journal of the Eastern Asia Society for Transportation Studies, Vol. 6, pp. 1123 - 1136, 2005
1136