1 Introduction
2 Brief literature review
2.1 Rolling resistance models
-
When the rolling resistance coefficient increases, the vehicle fuel consumption increases significantly, especially on roads with no gradient and at constant speed (typically high highway speed) (Bendtsen 2004).
-
The most significant pavement parameters affecting rolling resistance are macrotexture (MPD), or megatexture, unevenness or roughness (IRI) and stiffness.
-
Texture and unevenness affect the rolling resistance in a negative way; greater values of MPD and IRI correspond to greater rolling resistance.
-
For light vehicles, the impact of MPD is around three times that of the IRI effect.
-
The effect of roughness on rolling resistance can change with speed, while that of texture does not.
-
How stiffness affects PVI has not been consistently explained and is as yet, uncertain.
2.2 LCA studies including the rolling resistance component
Study | Country | Rolling resistance components included | Comments |
---|---|---|---|
Santero and Horvath (2009) | USA | Roughness | Rough estimation based on literature data |
Zhang et al. (2009) | USA | Roughness | Simple linear relationship between IRI and fuel consumption based on data from heavy duty trucks only, tested at low speed on test track |
Wang et al. (2012b) | USA | Roughness and texture | HDM-4 was used to consider the rolling resistance and MOVES (Motor Vehicle Emission Simulator) (EPA’s Office of Transportation and Air Quality (OTAQ) 2014) was used to model the vehicle emissions as a function of rolling resistance |
Yang (2014) | USA | Roughness | Model presented by Zaabar and Chatti (2010) |
Santos et al. (2015) | Portugual | Roughness and texture | Model presented by Hammarström et al. (2012) |
Bryce et al. (2014) | USA | Roughness and texture | |
Araújo et al. (2014) | Portugual | – | The energy consumption variation associated with different rolling resistances of the surface layers is evaluated with laboratory tests |
Wang et al. (2014) | USA | Roughness and texture | The vehicle CO2 emission factors are estimated as a continuous function of MPD and IRI, by using HDM-4 and MOVES (Motor Vehicle Emission Simulator) |
Xu et al. (2015) | USA | Roughness | Model presented by Zaabar and Chatti (2010) |
2.3 Parameters affecting the results of the rolling resistance component in LCA studies
2.4 Case study
Results for the base case scenario | |
---|---|
LCA phase | Tonne |
2009 reconstruction | 370 CO2e |
2009 traffic delay for the work-zone | 1.94–16.46 CO2
|
2009–2029 use phase (rolling resistance due to pavement surface MPD and IRI) | This paper |
3 Methodology
3.1 Calculation of the tailpipe CO2 emissions with VTI and UCPRC model
3.2 Sensitivity test
-
Traffic growth model
-
Pavement deterioration
-
‘average’ deterioration scenario (IRI increases from 1.0 to 2.3 m/km and MPD decreases from 1.8 to 0.8 mm);
-
‘worst’ deterioration scenario (IRI increases from 1.0 to 5.0 m/km and MPD is 1.5 mm during all the analysis period).
-
‘no deterioration’ scenario where the surface pavement condition is unchanged over time (IRI = 1.0 m/km and MPD = 1.5 mm).
-
Emission factor or fuel efficiency improvement
4 Results
Case scenario | Pavement deterioration | Traffic growth | Fuel efficiency /emission factors | Comments |
---|---|---|---|---|
Base case scenario | Average | No | No | Comparison of rolling resistance models |
Average case scenario | Average | Average | Average | Comparison of sensitivity test |
4.1 Comparison of the CO2 emissions calculated with the VTI and UCPRC models
Results for the base case scenario | ||||
---|---|---|---|---|
LCA phase | Result | |||
2009 reconstruction | 370 tCO2e | |||
2009 traffic delay for the work-zone | 1.94–16.46 tCO2
| |||
Total emissions | Basic component | Deterioration component | ||
2009–2029 use phase (rolling resistance due to pavement surface MPD and IRI) | UCPRC model | 1387 tCO2
| 1170 tCO2
| 217 tCO2
|
VTI model | 9672 tCO2
| 10,272 tCO2
| −600 tCO2
|
4.2 Sensitivity analysis
-
for both models, the potential emissions due to PVI rolling resistance have a large range of values;
-
this is particularly so in the deterioration component, especially in the VTI model, where the CO2 emissions can vary between 0.80 and 7.38 times the average value;
-
the best case scenario (lowest emissions) occurs under different assumptions for the two models (no deterioration in the UCPRC model and average deterioration in the VTI model). In the UCPRC model, the deterioration component increases over time, so the absence of deterioration minimizes the total emissions. In the VTI model, the deterioration component, under the average condition of pavement deterioration, tends to decrease, producing an overall reduction in the calculated emissions. This effect levels off under the ‘worst deterioration’ pavement condition, when the IRI effect is larger than the MPD effect.
Case scenario | Sensitivity parameter | Emission of CO2 (tonne) | ||||
---|---|---|---|---|---|---|
Pavement deterioration | Traffic growth | Emission factors | Basic | Deterioration | Total | |
Average case scenario | Average deterioration | Average | Average | 1288 | 225 | 1513 |
Best case scenario | No pavement deterioration | No | Average + 10 % | (−21 %) 1020 | (−100 %) 0 | (−33 %) 1020 |
Worst case scenario | Worst deterioration | Average + 10 % | No | (+36 %) 1755 | (+438 %) 1210 | (+96 %) 2965 |
Case scenario | Sensitivity parameter | Emission of CO2 (tonne) | ||||
---|---|---|---|---|---|---|
Pavement deterioration | Traffic growth | Emission factors | Basic | Deterioration | Total | |
Average case scenario | Average deterioration | Average | Average | 10,372 | −514 | 9858 |
Best case scenario | Average deterioration | No | Average + 10 % | (−12 %) 9141 | (+8 %) −557 | (13 %) 8584 |
Worst case scenario | Worst deterioration | Average + 10 % | No | (−1 %) 10,272 | (−738 %) 3281 | (+37 %) 13,553 |