Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
D-1 APPENDIX D FIELD ESTIMATES OF TRUCK WEIGHT-TO-POWER RATIOS A key issue in need of resolution is the distribution of truck characteristics related to performance on upgrades, particu- larly truck weight-to-power ratios. For over 30 years, from the 1950s through the 1980s, truck weight-to-power ratios decreased dramatically as truck engines became more and more powerful. Trucks with weight-to-power ratios greater than 150 kg/kW [250 lb/hp] have largely disappeared, with the exception of certain bulk haul operations like coal trucks and trucks hauling construction materials. Limited data sug- gest the distribution of truck weight-to-power ratios may have changed only a little since the mid-1980s, but this is uncertain. For the 2001 Green Book, a truck with a weight-to- power ratio of 120 kg/kW [200 lb/hp] was chosen as the basis for computation of critical length of grade. This seems a rea- sonable choice, but the data currently available could also support a choice in the 90- to 108-kg/kW [150- to 180-lb/hp] range. Therefore, field data were collected to document the actual distribution of truck weight-to-power ratios in the current truck fleet. This appendix describes the alternative approaches that were considered to collect the needed data, the site selection process for data collection locations, the data collection procedures, the data reduction procedures, and the field study results. DATA COLLECTION APPROACH Two fundamentally different approaches were considered to collect field data on truck weight-to-power ratios. These are: ⢠Measure truck weights and record marked engine power (referred to as nameplate horsepower) at weigh scales ⢠Record crawl speeds of trucks near the top of extended grades whose geometric profiles are known In the first approach, weight-to-power ratio is computed directly as a ratio. The weight can be determined accurately from weigh scale data. The actual net horsepower (after mechanical losses) must be estimated from the nameplate horsepower. This approach requires substantial cooperation from weigh scale operators, because all trucks (including empty trucks) must be weighed to get a representative sam- ple; many weigh scale operators allow empty trucks to bypass the scales because there is little point in weighing them for enforcement purposes. Cooperation of truckers is needed to determine the engine characteristics of their vehicles. This data collection approach requires a substantial amount of time per truck measured. In the second approach, weight-to-power ratios are com- puted from the truck crawl speeds on steep upgrades measured with a radar or lidar device. This computation requires esti- mates of the frontal area of the truck and its aerodynamic drag coefficient, although for trucks with substantially reduced crawl speeds, aerodynamic drag is relatively unimportant. However, the field measurementâtruck speedâis also a quantity of direct interest in highway design, since critical length of grade is based on a speed reduction criterion. The speed measurement approach has the advantages that data can be gathered relatively inexpensively and that no cooper- ation from weigh scale operators or truckers is needed; the only permission needed is permission from a highway agency to conduct a speed study from the roadside. In previous research for the Korean Institute of Construc- tion Technology, both field data collection methods described above were shown to provide comparable results in terms of the distribution of weight-to-power ratios. Therefore, the speed measurement approach was employed because it is more efficient. SITE SELECTION Data collection locations were selected through a three- step process. It was decided to use both freeway and two-lane highway locations because there were expected to be poten- tial differences in the truck population between these two types of highways. In addition, it was decided to use sites in both eastern and western states to investigate differences in the truck population in different regions of the country. The first step in site selection was to obtain data that would aid in selection of appropriate states and sites. Highway Per- formance Monitoring System (HPMS) data were utilized for this purpose. Table D-1 presents the mileage of rural free- ways and two-lane highways with grades over 4.5 percent by region of the U.S. Listings of the mileage of grades over 4.5 percent by state were also obtained (15). Based on the HPMS data and the proximity of potential sites within a rea- sonable traveling distance, a choice was made to collect truck crawl speeds on sustained grades in California, Colorado, and Pennsylvania. The second step in site selection was to conduct site visits in the respective states to identify and review candidate data collection locations. A total of 15 freeway sites and 15 two-lane highway sites were visited. In addition, truck crawl speed data from one rural two-lane highway site in California were available from a 1997 field study for NCHRP Project 3-55 (3) and were used
in this study. The following criteria were used to evaluate the appropriateness of a grade for purposes of this study. The characteristics of an ideal site include: ⢠Steep grade (at least 4 percent or more) ⢠Long grade (at least 1 mile in length) ⢠No sharp horizontal curves ⢠Good observation locations ⢠Relatively constant grade Figures D-1 to D-4 provide illustrations of several candi- date data collection sites in Colorado and Pennsylvania. The final step in verifying the suitability of sites for data collection was to review the actual values of percent grade from vertical profiles in as-built construction plans or from HPMS data. The actual local percent grade is needed to com- pute truck weight-to-power ratio from the truck crawl speed. Copies of vertical profile sheets from as-built plans and HPMS data were obtained. Based upon a review of the ver- tical profiles, 10 sites were selected for data collection: four sites in Colorado, three sites in Pennsylvania, and three sites in California (including the site for which data were already available). Table D-2 summarizes the characteristics of each site. The last three columns of the table present the average grade, the local grade in the vicinity of the data collection location, and the length of the grade from the foot of the D-2 Figure D-1. I-70 WB in Colorado on the approach to the Eisenhower Tunnel. Figure D-2. U.S. 285 NB (Crow Hill) in Colorado. Figure D-3. I-80 WB in Pennsylvania. Figure D-4. State Route 26 SB in Pennsylvania. Length of rural freeway (mi) Length of rural two-lane highway (mi) Region 4.5-6.4% > 6.4% 4.5-6.4% > 6.4% Northeast 244.7 19.9 407.3 284.9 Southeast 34.6 9.0 539.6 239.2 Midwest 50.6 16.0 521.2 153.9 West 254.1 15.6 948.8 109.3 California 108.8 4.0 287.6 48.8 TOTAL 692.8 64.5 2,704.5 836.1 TABLE D-1 Roadway mileages with steep grades (15)
D-3 State Route MP/Segmen t Directi on Name of pass/area Type of site Dates of data collection Avera ge grade (%) Local grade (%) Length of grade (mi) CA I-80 Pla/51 EB Baxter Freeway 8/12/02 & 8/14/02 4.8 5.0 2.0 CA I-80 Nev/3.7 EB Donner Summit Freeway 8/13/02 4.2 3.8 4.1 CA Rt 97 Siskiyou County Two- Lane 6/9/97 & 6/10/97 4.3 4.3 CO I-70 MP 210.9 EB Eisenhower Tunnel Freeway 8/2/02 6.4 6.5 5.9 CO I-70 MP 215.8 WB Eisenhower Tunnel Freeway 8/5/02 & 8/6/02 4.0 6.5 11.7 CO US 50 MP 192.7 EB Monarch Pass Two- Lane 7/31/02 & 8/1/02 4.6 5.5 2.8 CO US 285 MP 224 NB Crow Hill Two-Lane 7/29/02 & 7/30/02 6.9 7.0 1.9 PA I-80 Segment 1505 WB Centre County Freeway 11/08/02 3 3 1.5 PA Rt 26 Segment 20 SB Centre County Two- Lane 10/11/02 & 11/11/02 6.3 7.44 1.3 PA Rt 153 Segment 730 NB Clearfield County Two- Lane 11/04/02 & 11/05/02 6.3 8 1.2 TABLE D-2 Site information on data collection locations grade to the data collection location. Several typical data col- lection sites are illustrated in Figures D-1 through D4. DATA COLLECTION PROCEDURES Speed data were collected near the top of each grade to ensure that all or most trucks were operating at crawl speeds (i.e., the maximum speed of which the truck is capable at that point on the grade). Speeds were measured using a lidar gun. Speeds were recorded to the nearest mile per hour, and the configuration or type of each truck was recorded as follows: ⢠Single unit ⢠Single unit bulk carrier ⢠Van semitrailer ⢠Flat bed semitrailer ⢠Bulk semitrailer ⢠Low-boy semitrailer ⢠Auto carrier semitrailer ⢠Tank semitrailer ⢠Log semitrailer ⢠Single unit truck with trailer ⢠Double-trailer combination ⢠Triple-trailer combination ⢠Other The goal was to collect the speeds of 100 trucks at each two-lane highway site and 400 trucks at each freeway site. Speeds were measured only for unimpeded trucks; if a truck was traveling behind another vehicle, the truckâs speed may have been limited by its leader, so the truck speed was not measured. The speeds of trucks traveling in the through travel lanes were recorded, but the speeds of trucks driving on the shoulders were not recorded. DATA REDUCTION AND ANALYSIS The field study database included the truck type and speeds for each truck measured. The site-specific distribution of truck speeds is, in itself, of interest in highway design, but the primary purpose for conducting the field studies was to estimate the distribution of truck weight-to-power ratios in the current truck fleet. Therefore, the speed data were con- verted into truck weight-to-power estimates. The conversion of truck speeds to weight-to-power ratios was accomplished using the vehicle performance equations from the TWOPAS model (85, 86). Appendix E presents a truck speed profile model (TSPM) based on the TWOPAS vehicle performance equations. The TSPM can estimate the speed profile for the unimpeded truck on any specified verti- cal alignment given the truckâs weight-to-power ratio. For the analysis of the field data, the TWOPAS model was applied in reverse to estimate the truck weight-to-power ratio that would have produced the observed truck crawl speed on the known vertical alignment.
D-4 Number of observations 1,195 25th percentile ratio 112 Median ratio 141 75th percentile ratio 164 85th percentile ratio 183 90th percentile ratio 198 Weight-to-Power Ratio (lb/hp) Figure D-5. Distribution of estimated weight-to-power ratios for California freeways. RESULTS The distributions of weight-to-power ratios, categorized by type of site and by state, are presented below. Freeways Figures D-5, D-6, and D-7 show the distributions of weight- to-power ratios found for California, Colorado, and Pennsyl- vania freeways, respectively. The cumulative percentiles (by state) are presented in Table D-3. The 85th-percentile weight- to-power ratio from each state ranged from a low of 101 kg/kW [169 lb/hp] in Colorado to a high of 124 kg/kW [207 lb/hp] in Pennsylvania. Two-Lane Highways Figures D-8, D-9, and D-10 show the distributions of weight-to-power ratios found on California, Colorado, and Pennsylvania two-lane highways, respectively. The cumula- tive percentiles (by state) are presented in Table D-4. The 85th-percentile weight-to-power ratio from each state ranged from a low of 108 kg/kW [180 lb/hp] in Colorado to a high of 168 kg/kW [280 lb/hp] in Pennsylvania. Speed Distributions Table D-5 presents a summary of the speed distributions as measured at the respective data collection locations. In general, truck crawl speeds were greater on the freeways as compared to the two-lane highways, as would be expected due to the character of service they are intended to provide and the higher criteria to which they are designed. SUMMARY OF RESULTS It is difficult to decide whether the truck populations found in the three states differ by state or geographic region (east- ern states vs. western states). However, it is clear from the results obtained that the truck population on two-lane high- ways generally has greater weight-to-power ratios than the truck population on freeways. Clearly, the best performing truck fleet is in Colorado, and the poorest performing truck fleet is in Pennsylvania. Fur- ther, there is much more variability in truck weight-to-power ratios on two-lane highways than on freeways. Long-haul trucks may have the best weight-to-power ratios, and one would expect the long-haul trucks to be more prevalent on the freeways and less so on two-lane highways. In summary, the 85th-percentile weight-to-power ratio on freeways falls within a fairly narrow range, from 102 to 126 kg/kW [170 to 210 lb/hp] nationally, with California and Colorado at the low end of that range and Pennsylva- nia at the high end. For design purposes, it appears that a truck with weight-to-power ratio of 102 to 108 kg/kW [170 to 180 lb/hp] would be appropriate for freeways in Cali- fornia and Colorado, while a weight-to-power ratio of 126 kg/kW [210 lb/hp] would be more appropriate in
D-5 Number of observations 734 25th percentile ratio 87 Median ratio 115 75th percentile ratio 152 85th percentile ratio 169 90th percentile ratio 179 Weight-to-Power Ratio (lb/hp) Figure D-6. Distribution of estimated weight-to-power ratios for Colorado freeways. Number of observations 431 25th percentile ratio 142 Median ratio 168 75th percentile ratio 194 85th percentile ratio 207 90th percentile ratio 220 Weight-to-Power Ratio (lb/hp) Figure D-7. Distribution of estimated weight-to-power ratios for Pennsylvania freeways.
D-6 Weight-to-power (lb/hp) ratio Percentile California Colorado Pennsylvania 5th 83 69 111 25th 112 87 142 50th 141 115 168 75th 164 152 194 85th 183 169 207 90th 198 179 220 95th 224 199 251 TABLE D-3 Summary of truck weight-to-power ratios for freeway sites Number of observations 250 25th percentile ratio 144 Median ratio 186 75th percentile ratio 226 85th percentile ratio 246 90th percentile ratio 262 Weight-to-Power Ratio (lb/hp) Figure D-8. Distribution of weight-to-power ratios for California two-lane highways. Number of observations 264 25th percentile ratio 86 Median ratio 107 75th percentile ratio 150 85th percentile ratio 180 90th percentile ratio 193 Weight-to-Power Ratio (lb/hp) Figure D-9. Distribution of estimated weight-to-power ratios for Colorado two-lane highways.
Pennsylvania. For two-lane highways, a 108-kg/kW [180- lb/hp] design vehicle may be appropriate in Colorado, while less powerful design vehicles in the 150 to 168 kg/kW [250 to 280 lb/hp] range may be appropriate for California and Pennsylvania. All of these design vehicle weight-to-power ratios represent the 85th percentile of the truck population; so, of course, most of the truck popula- tion performs substantially better. D-7 Number of observations 297 25th percentile ratio 110 Median ratio 180 75th percentile ratio 242 85th percentile ratio 280 90th percentile ratio 303 Weight-to-Power Ratio (lb/hp) Figure D-10. Distribution of weight-to-power ratios for Pennsylvania two-lane highways. Weight-to-Power (lb/hp) Ratio Percentile California Colorado Pennsylvania 5th 79 68 79 25th 144 86 110 50th 186 107 180 75th 226 149 242 85th 246 180 280 90th 262 193 303 95th 281 214 331 TABLE D-4 Summary of truck weight-to-power ratios for two-lane highway sites Speed (mph) State Route MP/Segment Directi on Type of site Number of observations 15th percentile 50th percentile 85th percentile CA I-80 Pla/51 EB Freeway 600 29 36 52 CA I-80 Nev/3.7 EB Freeway 600 32 41 53 CA Rt 97 Two-Lane 250 25 32 52 CO I-70 MP 210.9 EB Freeway 400 26 39 54 CO I-70 MP 215.8 WB Freeway 350 26 36 50 CO US 50 MP 192.7 EB Two-Lane 97 27 41 50 CO US 285 MP 224 NB Two-Lane 169 22 40 51 PA I-80 Segment 1505 WB Freeway 434 38 45 55 PA Rt 26 Segment 20 SB Two-Lane 109 17 31 41 PA Rt 153 Segment 730 NB Two-Lane 189 12 19 33 TABLE D-5 Speed distributions at data collection locations
