AP-R211 GEOMETRIC DESIGN FOR TRUCKS WHEN, WHERE AND HOW? A USTROADS

AP-R211 GEOMETRIC DESIGN FOR TRUCKS WHEN, WHERE AND HOW? A USTROADS

First Published 2002 Austroads Inc. 2002 This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may be reproduced by any process without the prior written permission of Austroads. National Library of Australia Cataloguing-in-Publication data: Geometric Design for Trucks When, Where and How? ISBN 0 85588 632 3 Austroads Project No. T&E.D.N.009 Austroads Publication No. AP R211/02 Project Manager John Cunningham, VicRoads Prepared by John McLean (Consulting Engineer) Michael Tziotis, ARRB Transport Research Thanuja Gunatillake, ARRB Transport Research Published by Austroads Incorporated Level 9, Robell House 287 Elizabeth Street Sydney NSW 2000 Australia Phone: +61 2 9264 7088 Fax: +61 2 9264 1657 Email: austroads@austroads.com.au www.austroads.com.au Austroads believes this publication to be correct at the time of printing and does not accept responsibility for any consequences arising from the use of information herein. Readers should rely on their own skill and judgement to apply information to particular issues.

GEOMETRIC DESIGN FOR TRUCKS WHEN, WHERE AND HOW? Sydney 2002

AUSTROADS PROFILE Austroads is the association of Australian and New Zealand road transport and traffic authorities whose purpose is to contribute to the achievement of improved Australian and New Zealand transport related outcomes by: developing and promoting best practice for the safe and effective management and use of the road system providing professional support and advice to member organisations and national and international bodies acting as a common vehicle for national and international action fulfilling the role of the Australian Transport Council s Road Modal Group undertaking performance assessment and development of Australian and New Zealand standards developing and managing the National Strategic Research Program for roads and their use. Within this ambit, Austroads aims to provide strategic direction for the integrated development, management and operation of the Australian and New Zealand road system through the promotion of national uniformity and harmony, elimination of unnecessary duplication, and the identification and application of world best practice. AUSTROADS MEMBERSHIP Austroads membership comprises the six State and two Territory road transport and traffic authorities and the Commonwealth Department of Transport and Regional Services in Australia, the Australian Local Government Association and Transit New Zealand. It is governed by a council consisting of the chief executive officer (or an alternative senior executive officer) of each of its eleven member organisations: Roads and Traffic Authority New South Wales Roads Corporation Victoria Department of Main Roads Queensland Main Roads Western Australia Transport South Australia Department of Infrastructure, Energy and Resources Tasmania Department of Infrastructure, Planning and Environment Northern Territory Department of Urban Services Australian Capital Territory Commonwealth Department of Transport and Regional Services Australian Local Government Association Transit New Zealand The success of Austroads is derived from the synergies of interest and participation of member organisations and others in the road industry.

Executive Summary Background While truck tracking characteristics are a primary consideration for intersection geometry and curve widening practices, most road geometric design standards, particularly alignment design standards, are based on the performance characteristics of passenger cars and car drivers. Austroads Project T&EDN 009 was aimed at deriving truck-based geometric design standards and establishing the traffic volumes and mix at which the adoption of such standards is economically warranted. Design Truck Characteristics Design truck performance characteristics were established from an assessment of the literature and cooperation with an Austroads/NRTC project aimed at developing performance-based standards for the regulation of heavy vehicles. The exercise focussed on the characteristics of the 6-axle semi-trailer. This is typically the most common truck type on inter-regional freight routes and is representative of the largest trucks allowed general access to the road network. The proposed truck parameter values for geometric design are tabulated below. Proposed Truck Parameter Values for Geometric Design Independent of Design Speed. Parameter Value Mass (tonne) 42.5 Length (m) 19.0 P/M ratio (kw/tonne) 6.1 Driver eye height (m) 2.4 SRT 0.35 Proposed Truck Parameter Values for Geometric Design Dependent on Design Speed. Design Speed (km/h) Longitudinal Deceleration (g) Side Friction Factor (g) Desirable Absolute Max. 50 0.29 0.21 0.25 60 0.29 0.17 0.24 70 0.29 0.14 0.23 80 0.29 0.13 0.20 90 0.29 0.12 0.15 100 0.28 0.12 0.12 110 0.26 0.12 0.12 120 0.25 0.11 0.11

Truck-Based Design Standards The report contains truck-based design standards, derived from the above values, for the following design elements: horizontal curve standards based on limiting lateral acceleration criteria; crest vertical curves based on stopping sight distance; horizontal curve radii and lateral clearances based on stopping sight distance; acceleration lane lengths; and vertical grades. The truck-based, crest vertical curve standards are less demanding than the car-based standards, indicating that the increased driver eye height for trucks more than compensates for the poorer deceleration performance, and that car-based crest standards will also meet the requirements for trucks. Acceleration lane lengths based on the criterion of a truck accelerating to the design speed of the through lanes are provided in the report. However, unless the acceleration lane can be combined with a downgrade, the required lengths are not practical. Additional lengths based on the criteria of trucks accelerating to 10 and 20 km/h less than the through lane speed are also provided. Similarly, grade standards based on a truck being able to maintain the design speed are provided, but these will often not be achievable in practice. Additional standards based on truck speeds not falling more than 10 and 20 km/h below the design speed are also provided. Economic Justification Benefit-cost analysis was undertaken for a number of examples to indicate the initial year truck volumes at which truck-based rather than car-based standards would be economically justified. The results were reasonably consistent for curve radius and horizontal sight distance standards. Depending on terrain conditions and road type (single-carriageway or dual-carriageway), truck-based standards could be justified at those volumes typically found on the higher volume freight routes. Indicative values are given in the report. The results for grade standards were highly variable, depending on the interaction between the road grade and the terrain. Further, the adoption of arbitrarily flat grade standards could result in an increase in route length, which would both increase road provision costs and, to a greater or lesser extent, offset the user benefits from the flatter grade. The road safety implications of truck performance on long, steep downgrades were also not considered in the analysis. Hence, it is recommended that the question of truck-based grade standards be considered on a case-by-case basis.

CONTENTS 1 INTRODUCTION...1 1.1 Purpose...1 1.2 Changes To Traffic Mix And Truck Characteristics...1 1.3 Current Practices...1 1.3.1 Australian...1 1.3.2 International...2 2 PROJECT METHODOLOGY...3 2.1 Approach...3 2.2 Equivalent Risk Concept...3 2.3 Austroads/NRTC Performance Based Standards Project...4 3 DESIGN VEHICLE...5 3.1 General...5 3.2 Acceleration and Gradeability...5 3.2.1 Design Vehicle Model...5 3.2.2 Design Vehicle Parameters...6 3.3 Roll Stability...6 3.4 Sight Distance Parameters...7 3.4.1 Deceleration...7 3.4.2 Eye and Object Height...8 3.4.3 Perception Reaction Time...9 4 DESIGN ELEMENTS...10 4.1 Grades...10 4.2 Acceleration Lanes...10 4.3 Horizontal Curves...11 4.4 Stopping Sight Distance Related...12 4.4.1 Stopping Distance...12 4.4.2 Crest Vertical Curves...13 4.4.3 Horizontal Curves...13 5 TRUCK-BASED DESIGN VALUES...14 5.1 Grades...14 5.2 Acceleration Lanes...14 5.3 Minimum Horizontal Curve Radius...15 5.4 Standards Based on Stopping Sight Distance...16 5.4.1 Crest Vertical Curves...16 5.4.2 Sight Distance on Horizontal Curves...16 6 WHEN TO APPLY TRUCK-BASED STANDARDS...18 6.1 Benefit-Cost Analyses...18 6.2 Terrain...18 6.3 Curve Radius and Horizontal Sight Distance...18 6.4 Grade...19 7 REFERENCES...20

1 INTRODUCTION 1.1 Purpose While truck tracking characteristics are a primary consideration for intersection geometry and curve widening practices, most road geometric design standards, particularly alignment design standards, are based on the performance characteristics of passenger cars and car drivers. Austroads Project T&EDN 009 was aimed at deriving truck-based geometric design standards and establishing the traffic volumes and mix at which the adoption of such standards is economically warranted. The current document represents the final report for the project. More detailed information is given in three previous progress reports. The first progress report (McLean and Newman 2001) documented a review of local and international information and practices related to truck-based design standards. The second progress report (McLean 2001a) documented the bases of proposed truck performance parameter values, and the truck-based geometric design standards for design elements derived from them. The third progress report (McLean 2001b) documented the development of simple user benefit models for truck-based standards and investigations of truck and mixed traffic volumes for which the adoption of truck-based standards could be economically warranted. The present report summarises the findings in a single document. 1.2 Changes to Traffic Mix and Truck Characteristics Extensive research into alignment design standards for Australian roads was undertaken some 25 years ago. At that time articulated trucks typically travelled at about 80 % of the speed for cars. Because of this speed differential it was considered that stopping sight distance and curve standards based on the 85 th percentile speed car would generally be adequate for trucks. With the standards for the design elements being proportional to speed squared, the speed difference would more than compensate for the poorer braking and cornering performance of trucks relative to cars. Since that time there has been a near continuous trend of increasing power for truck engines, improved aerodynamic design, and the adoption of lower rolling resistance tyres for trucks. These have all contributed to a general increase in travel speeds for trucks. For open road conditions, the mean speed of trucks is currently very close to that for cars and the difference between truck and car 85 th percentile speeds is less than 5 %. Therefore, we can no longer assume that slower travel speeds for trucks will compensate for poorer braking and cornering performance. There has also been a marked change in the mix of traffic. Truck proportions were typically 10 %, with 20 % regarded as a high truck proportion, and 2-axle rigid trucks were the dominant truck type. Truck proportions of 30 to 50 % are now common on major inter-regional freight routes, with most of the trucks being the larger articulated heavy vehicles. Total travel by trucks has been growing at about twice the growth in car travel and recent traffic growth estimates by the BTE suggest that this is likely to continue for at least another 15 years (Gargett et al 1999). 1.3 Current Practices 1.3.1 Australian The extra length and swept width demands of large trucks are a critical factor for intersection design. For reasons described above, trucks have not traditionally been considered in the specification of standards for alignment design elements. A number of recent State agency guides, and the current draft Austroads guide, provide alternative standards for some design elements based on relevant truck performance characteristics. These appear to be based on very conservative design criteria, resulting in standards which would be difficult to apply in other than very easy terrain. Further, while the truck-based standards are provided as an alternative, there is no guidance on the circumstances in which they should be adopted. Hence there is a reluctance to employ the truck-based standards in practice. 1

1.3.2 International The US Surface Transportation Assistance Act of 1982 increased the mass and dimensional limits for articulated trucks operating on a designated National Network of highways. Further, the States were required to provide access between the National Network and terminals or services. Because of concerns that the new truck limits had not been considered in the development of the, then, current AASHTO geometric design policy, extensive investigations were undertaken on the relationships between truck dimensional and performance characteristics and the standards for geometric design elements (see, for example, Transportation Research Board 1986). While absolute relationships could be developed between the truck dimensions and intersection geometry, the relationships between truck performance and the standards for alignment design elements were more relative. A new design truck reflecting the changes in dimensional limits was specified for intersection design but the existing car-based standards were retained for alignment design elements. In interpreting this development in the Australian context it should be noted that, for design speeds of 90 km/h or less, the US car-based standards are generally more conservative than those employed in Australia. A number of European countries revised their design standards in the early 1990s in anticipation of a new road construction program to facilitate East-West trade. Some background on these revisions is given in Texas Transportation Institute (1998). The width requirements of trucks have clearly been a consideration in the revision of width standards and truck gradeability is considered in grade standards. There is no evidence that the cornering or braking performance of trucks has been considered in the development of relevant alignment design standards. However, European countries typically impose an 80 km/h speed limit on trucks so that, at least on the higher design speed roads, a speed differential will compensate for the poorer braking and cornering performance of trucks relative to cars. 2

2 PROJECT METHODOLOGY 2.1 Approach The project was conducted in four parts. 1. Literature Review: A desktop evaluation was undertaken of a number of Australian and overseas design guides and the relevant research literature to establish the extent to which truck characteristics are considered in the specification of design standards. 2. Defining the Design Truck: Representative truck performance characteristics relevant to the derivation of truck-based standards for alignment design elements were specified. These covered gradeability and acceleration, cornering stability, and braking ability. This part of the project involved liaison with the Austroads/NRTC performance based standards project described below. 3. Derivation of Truck-Based Standards: The truck performance characteristics were employed in standard design models to derive truck-based standards for grade, acceleration lane length, horizontal curves, stopping sight distance, and sight distance constraining design elements. As the stopping sight distance and horizontal curve design criteria values for cars carry an implicit level of acceptable risk, the equivalent risk concept described below was employed to apply the same level of risk to the design values for trucks. 4. Benefit-Cost Analysis: Abstracted models of design elements and terrain were developed to estimate the costs of providing truck-based rather than car-based standards for individual design elements. Simple user cost models were employed to estimate the benefits to users from the provision of truck-based rather than car-based standards. These were combined in a benefit-cost analysis to provide coarse estimates of the traffic or truck volumes at which the adoption of truck-based would be economically justified. 2.2 Equivalent Risk Concept While safety is a primary consideration for geometric design, standards for many design elements are based on an implied acceptable level of risk rather than relationships between design standard and crash frequency. For example, the design model used to derive standards for elements based on stopping sight distance defines a performance envelope which is considered to provide an acceptable level of risk from sight distance related crashes taking into account travel speed, sight geometry, perception-reaction time, and achievable deceleration rates. However, the safety implications for vehicles operating outside of the performance envelope are not clear (Glennon 1987). This does not necessarily mean that there is no relationship between sight distance and crash propensity, but that it has not been possible to extract such a relationship from the range of confounding factors which contribute to sight distance related crashes. The most recent research from the USA (Fambro et al 1997) suggests that a relationship may exist, but that it is non-linear. To avoid the difficulties of relating safety to geometric design criteria, an equivalent risk approach was adopted for the current project. The equivalent risk concept was employed both for establishing truck-based design criteria values (eg, truck-based side friction factor design values for horizontal curves) and, for example, benefit-cost analyses of the truck volumes at which the adoption of truck-based standards is economically justified. For the former case, the equivalent risk approach meant applying the same factor of safety between the truck-based design criteria values and the relevant design truck performance measure as is implicit in the car-based standards. For example, the ratio of the truck-based side friction factor design values to the static roll threshold of the design truck was taken to be the same as the ratio of the car-based side friction factor values and the actual friction that can be expected. The equivalent risk approach was employed in the benefit-cost analysis for those cases where the main user cost effect from variation in a geometric standard was expected to be changes in crash costs. In the absence of relationships between the value for the standard and crash frequency, the effect on crash costs could not be calculated. The approach employed was to infer user cost differences by varying a quantifiable operating parameter (particularly speed) for trucks to operate at the same level of risk as is currently implied for cars. 3

For example, trucks typically negotiate minimum sight distance elements at speeds less than cars, but greater than the speed defined by the stopping sight distance performance envelope. Under the inferred cost approach, it was assumed that trucks negotiate such features at the speed given by the performance envelope, and the differences in speed with geometric standard was used to infer equivalent risk differences in truck operating costs based on travel time. 2.3 Austroads/NRTC Performance Based Standards Project Austroads and the NRTC are managing a program of work aimed at providing a performance based standards (PBS) regime to regulate the performance characteristics of heavy vehicles currently controlled by mass and dimension regulations. The current project in this program (referred to as PBS A3/A4) is aimed at establishing appropriate performance standards, including performance levels. This will include an assessment of the performance characteristics of the current Australian heavy vehicle fleet. Many of the truck characteristics relevant to the specification of design vehicles are being investigated by PBS A3/A4 and close liaison was maintained for the specification of the design truck. 4

3 DESIGN VEHICLE 3.1 General The characteristics of the current heavy vehicle fleet are regulated by a set of mass and dimensional regulations, Australian Design Rules (ADRs), and Australian Vehicle Standards Regulations (AVSRs). Working within the regulatory controls, truck operators and designers attempt to optimise the vehicle for the freight task being performed. Hence, for a given vehicle configuration (semi-trailer, B-double, etc), the road design relevant vehicle characteristics tend to group into three clusters depending on the type of freight being carried. Vehicles carrying high density freight, so that mass regulations control the vehicle load. Vehicles carrying low density freight, so that dimensional regulations control the vehicle load. Tankers. While there are particular outlier vehicles, these tend to be specialised vehicles driven by aware drivers who compensate for the performance limitations. For example, remote area stock crates and urban garbage collection trucks have roll stability characteristics which are poor relative to more typical trucks. It had been hoped to draw on the results of a heavy vehicle fleet characteristics mapping exercise being undertaken by the PBS A3/A4 project (see Section 2.3) to establish relevant truck performance characteristics for each of the above three vehicle groups. However the timing of the two projects did not allow this. Instead, more general information used by the PBS A3/A4 project to develop recommended performance-based standards for heavy vehicle regulation was employed to establish more generic truck performance characteristics. In undertaking this exercise there was a particular focus on the 6-axle semitrailer. This vehicle is representative of the largest trucks allowed general access to the road network and is typically the most common vehicle type on inter-regional freight routes. 3.2 Acceleration and Gradeability 3.2.1 Design Vehicle Model Very sophisticated mathematical relationships and computer models are available for the assessment of individual vehicles, using manufacturer supplied technical specifications for the engine, transmission, and tyres as input. As road design and traffic analysis is concerned with representative vehicles, a simpler formulation is typically adopted with the engine and transmission assumed to be capable of delivering uniform power to the drive wheels over the operating speed range of interest (McLean 1989, Watanatada et al 1987). With this formulation, the acceleration performance equation is, dv/dt = P DR /(Mv) - 0.5ρC D Av 2 /M - (C R + θ)g (1) For sustained speed on a grade, dv/dt = 0, and (0.5ρC D A/M)v 3 + (C R + θ)gv - P DR /M = 0 (2) where: v = vehicle velocity (m/s) P DR = power delivered to the drive wheels (W) M = mass of the vehicle (kg) ρ = air density (1.22 kg/m 3 ) C D = aerodynamic drag coefficient A = projected frontal area (m 2 ) C R = coefficient of rolling resistance θ = gradient (m/m) g = acceleration due to `gravity (9.8 m/s 2 ) 5

3.2.2 Design Vehicle Parameters Botterill (1997) updated earlier estimates of representative air resistance and rolling resistance parameter values to reflect trends in aerodynamic design and tyre rolling resistance. These are given in Table 1. Table 1. Botterill (1997) estimates of typical air and rolling resistance parameters Parameter Heavy Rigid Truck Articulated Truck C D 0.6 0.65 A (m 2 ) 5.5 8.5 C R 0.014 0.010 Table 2 gives P/M values calculated from the proposed PBS A3/A4 gradeability standard and values recommended for design purposes. The latter are based on advice from the PBS A3/A4 project team on typical power/mass characteristics for different truck types. Source Table 2. Estimates of Representative Acceleration and Gradeability Performance Parameter Values PBS A3/A4 Gradeability standard (Backcalculated) M (t) 6-Axle Semi P/M (kw/t) 9-Axle B-Double M (t) P/M (kw/t) 42.5 5.2 62.5 4.2 Recommended for Design 42.5 6.1 62.5 4.8 The PBS gradeability standards are based on existing regulations which are aimed at setting a lower bound on vehicle performance. Hence the P/M ratios calculated from them may not be appropriate values for design vehicles, particularly given the industry trend to more powerful engines over the last decade. The recommended values are based on advice from the PBS A3/A4 project using information on the characteristics of the typical combination vehicles. 3.3 Roll Stability Horizontal curve standards are derived from the mechanics of circular motion, with friction and superelevation providing the centripetal force necessary to balance the inertial force. For an exact circular path and with small angle approximations, e + f = V 2 /127 R (3) where: e = curve superelevation (m/m) f = side friction factor (SFF) V = vehicle speed (km/h) R = curve radius (m) Trucks have a higher centre of gravity than cars. Consequently, the limiting condition for trucks negotiating a circular curve tends to be rollover rather than skidding. Truck roll stability is commonly expressed as a static roll threshold (SRT), with units of g (acceleration due to gravity). This is the component of lateral acceleration parallel to the road surface corresponding to the onset of rollover. Mathematically, SRT is equivalent to the SFF in the standard curve design equation, and limiting design values of SFF related to truck rollover can be employed in an equivalent way to the current values being related to risk of skidding. 6

The SRT for loaded large trucks typically ranges from about 0.30 to 0.5 g. In extreme cases, such as stock crate trucks operating in the remote regions of Australia, it can be less than 0.2. Ervin et al (1986) estimated SRT values for semi-trailers carrying full loads at different freight density. Trucks carrying medium density and high density freight gross out at less than cubic capacity, resulting in low centre of gravity and high SRT. Trucks carrying low density freight use all the cubic capacity available, resulting in higher centre of gravity and lower SRT values. The worst case is a flat deck truck carrying homogenous freight at a density that meets both gross mass limits and cubic capacity, resulting in an SRT of about 0.25. Various trends have served to improve SRT, particularly for higher productivity vehicles. There is a trend to smaller diameter truck tyres and the use of drop deck trailers by operators carrying low density freight. Further, some line-haul operators aim to carry a mix of high density and low density freight with a view to maximising the utilisation of both mass and cubic limits, with the low density freight being stacked on top of the high density freight. Based on studies undertaken in North America and New Zealand, the PBS A3/A4 project team has proposed an SRT of 0.35 g as the performance standard for heavy vehicle roll stability. This is considered to be an appropriate value of SRT for the design truck. 3.4 Sight Distance Parameters 3.4.1 Deceleration The Australian Vehicle Standards Rules 1999 (AVSRs) specify minimum standards for braking performance outcomes. For heavy vehicles ( 2.5 tonnes) the AVSRs specify a minimum average deceleration achievable in braking from any speed at which the vehicle can operate of 0.29 g and a peak deceleration achievable of 0.59 g. Note that the AVSR deceleration rates relate to the performance of the braking system on a dry surface rather than the available tyre/road friction which is assumed for car-based stopping sight distance standards. While the AVSR minimum braking performance standard for heavy vehicles may be a vehicle design control for large combination vehicles, it is likely to underestimate the real braking performance of rigid trucks and semi-trailers. Truck braking distances and stopping sight distance standards received research attention in the USA during the 1980s. Fancher (1986) developed theoretical wet surface stopping distance relationships for semi-trailers based on actual measurement of truck tyre skid resistance. The achievable decelerations given by these relations ranged between about 0.23g to 0.3g at 32 km/h and 0.17g to 0.23g at 96 km/h. These values formed the basis for truck stopping sight distance standards recommended by Harwood et al (1989). Harwood et al presented a range of truck deceleration values for a poor quality wet surface, with 0.17 representing a very conservative, worst-case condition. The upper bound ranged from 0.25g to 0.28g, depending on design speed. The lower bound value was adopted for truck stopping sight distance standards in the VicRoads guide (VicRoads 1996) and are proposed as interim values in the new Draft Austroads Guide. An examination of the Fancher relationships reveals that the achievable decelerations were controlled by the assumed surface friction properties rather than the braking system performance. The friction properties were based on a 60 km/h friction value of 0.28, which is very low by Australian standards, and certainly well below the available friction assumed in the derivation of the Austroads (1989) car-based stopping sight distance standards. Mason and Briggs (1985) derived a theoretical stopping distance versus initial speed relation for dry conditions from vehicle regulatory considerations, and compared it with the results of both dry and wet braking tests on combination vehicles. These tests were conducted at speeds up to 64 km/h, and the results of all except one were within the theoretical dry surface curve, corresponding to a deceleration of 0.37 g. 7

The Fancher (1986) truck stopping distance relationships were aimed at generalising the results of stopping distance tests by Olson et al (1984). These tests found that, for a given surface condition, controlled stops by semi-trailers typically required about 1.4 times the distance required by cars. As stopping distance is inversely proportional to average deceleration, the simplest way of applying the Olsen result is to reduce the deceleration rates assumed for cars by a factor of 1.4. This will provide the same level of risk within truckbased stopping sight distance standards as is currently provided for cars in the car-based standards, without the need to assume explicit values for skid resistance. It is proposed that truck deceleration design values be taken as the lesser of, the AVSR heavy vehicle minimum deceleration of 0.29 g, representing a control limit on the performance of the braking system; and the assumed car deceleration rate divided by 1.4, representing limits imposed by available friction. Values derived in this way are given in Table 3. The current car-based values are shown for comparison. Note that, in the absence of an articulation point, control under braking is of less concern for rigid trucks than for semi-trailers. Hence, the friction controlled deceleration values (design speed of 100 km/h or greater) are likely to be more conservative for rigid trucks than for semi-trailers. Also the AVSR controlled deceleration rates (design speeds of 90 km/h or less) are likely to be more conservative for the lighter classes of rigid trucks than for semi-trailers. Table 3. Proposed Longitudinal Deceleration Values for Truck Stopping Sight Distance Initial Speed (km/h) Proposed Semi-Trailer Longitudinal Deceleration (g) Austroads Car 50 0.29 0.52 60 0.29 0.48 70 0.29 0.45 80 0.29 0.43 90 0.29 0.41 100 0.28 0.39 110 0.26 0.37 3.4.2 Eye and Object Height Fitzpatrick Lienau and Fambro (1998) gives the results of driver eye height surveys undertaken as part of the NCHRP study of stopping sight distance. For heavy trucks, the mean driver eye height was 2.45 m and the 15 th percentile value was 2.34 m. Given the global nature of the heavy truck market, it can be expected that these values will also be representative of heavy trucks operating in Australia. The mean value was slightly higher than the value of 2.36 m measured some 15 years earlier (Olson et al 1984) and suggests that, unlike cars, driver eye heights suitable for heavy trucks are relatively stable and may be slowly increasing. The above result will be applicable to both semi-trailers and the heavier rigid trucks which have similar cab designs. Lower eye heights can be expected for the lighter classes of rigid trucks. However, as discussed above, the proposed longitudinal deceleration values will be more conservative for light rigid trucks than for semi-trailers. In terms of the overall stopping sight distance model, this should largely offset the effects of the lower driver eye heights for the lighter rigid trucks. A design value of 2.4 m is proposed. There is no rationale for adopting object height values for truck stopping sight distance different to those assumed for cars. 8

3.4.3 Perception - Reaction Time There has been some debate in the literature as to whether the perception-reaction time for calculating truck stopping sight distance should be reduced to allow for an experienced professional driver contribution (eg, Fancher 1986), but this needs to be balanced against concerns regarding driver fatigue and the delay of about 0.5 s inherent in air braking systems. The emerging consensus is that the perception-reaction time assumed for truck braking should be the same as that assumed for cars (Harwood et al 1989). The draft Austroads Guide recommends 2.5 s as the normal value for rural design, with an absolute minimum value of 2.0 s in constrained situations. It is proposed that the same values be adopted for truckbased standards. 9

4 DESIGN ELEMENTS 4.1 Grades Car gradeability is not of primary concern in current design standards for vertical grade that are derived primarily from level-of-service considerations and the discrepancy in upgrade speeds for cars and heavy vehicles. However, gradeability (the speed a vehicle can maintain on a sustained upgrade) is the primary consideration for truck-based grade standards as this impacts both on the operating costs for trucks and the congestion imposed on light vehicles by slow moving trucks. Referring to the gradeability equation (eqn 2), as air and rolling resistance are largely fixed by truck size and tyre technology, truck gradeability is mainly determined by the power/mass ratio. Truck-based grade standards can be derived using the design criterion of the truck being able to maintain the design speed on sustained upgrades. However, the grade standards necessary to meet this criterion are unrealistic for the higher range of design speeds and would be unachievable for the lower range of design speeds in other than flat terrain. Design criteria based on limiting the extent to which truck upgrade speeds fall below the design speed are proposed as being both more realistic and achievable. Speed decrements of 10 to 20 km/h are reasonable for the type of road and traffic volumes at which truck-based standards might be considered. While the design speed or speed decrement criteria would ideally be applied through an analysis of truck speed along the vertical profile, the essence of the grade requirements which satisfy the criteria can be captured in two grade standards. 1. General maximum grade: This is the sustained grade at which the truck can maintain a speed corresponding to the design criterion. It can be calculated from the gradeability equation (eqn 2). 2. Length limits on steeper grades: For grades steeper than the general maximum, trucks will decelerate to a sustainable speed less than the speed criterion. The length limit represents the length of grade at which the truck will decelerate to the speed criterion value. It can be calculated through iterative application of the acceleration equation (eqn 1). 4.2 Acceleration Lanes Acceleration lanes are provided to enable entering traffic to accelerate to the design speed of the through roadway. Truck acceleration is largely determined by the power/mass ratio and, because of the poor acceleration performance of trucks relative to cars, trucks require a much greater distance to accelerate to a given speed. Figure 1 gives speed-distance plots for trucks in acceleration as derived from iterative application of the acceleration equation (eqn 1). 10

100 80 0 % Grade 100 80 1 % Upgrade Speed (km/h) 60 40 Speed (km/h) 60 40 20 20 0 10 100 1000 10000 Distance (m) 0 10 100 1000 10000 Distance (m) 100 80 2 % Upgrade 100 80 1 % Downgrade Speed (km/h) 60 40 Speed (km/h) 60 40 20 20 0 10 100 1000 10000 Distance (m) 0 10 100 1000 10000 Distance (m) 100 80 2 % Downgrade 100 80 4 % Downgrade Speed (km/h) 60 40 Speed (km/h) 60 40 20 20 0 10 100 1000 10000 Distance (m) 0 10 100 1000 10000 Distance (m) Fig. 1. Speed profiles for a semi-trailer starting from rest on constant grades. Unless the acceleration lane can be combined with a downgrade, the lengths of acceleration lane required for trucks to accelerate to the design speed of the through roadway are unrealistically long. Merging truck speeds 10 to 20 km/h less than the through speed would not be expected to be unduly disruptive to traffic flow. Acceleration lane lengths based on the truck accelerating to such speeds are more realistic and achievable. 4.3 Horizontal Curves As described in Section 3.3, the static roll threshold (SRT) of a truck is mathematically equivalent to the available tyre/road friction in the standard curve design equation. Truck-based design side friction factors (SFFs) can be derived by providing the same factors of safety between SFF and SRT as currently exist between car-based SFF and available friction. That is f T = SRT f C / f avail (4) where: f T = limiting value of side friction factor based on truck rollover SRT = static roll threshold for the design truck f C = current car-based limiting value of side friction factor f avail = tyre-road friction which can be expected to be available. 11

The VicRoads/RTA investigatory levels of skid resistance for SCRIM measurements (VicRoads and RTA 1995) provide lower range estimates for the values of f avail which can be expected in practice. Table 4 gives the values of f T derived in this way, with the car-based values from which they were derived shown for comparison. For design speeds of 100 km/h or greater available friction is the controlling criterion for both cars and trucks, and truck-based SFFs are the same as for cars. Table 4. Proposed Design Values of Truck Side Friction Factor V (km/h) ft Car-Based Design f Desirable Absolute Max. Desirable Absolute 50 0.21 0.25 0.30 0.35 60 0.17 0.24 0.24 0.33 70 0.14 0.23 0.19 0.31 80 0.13 0.20 0.16 0.26 90 0.12 0.15 0.13 0.18 100 0.12 0.12 0.12 0.12 110 0.12 0.12 0.12 0.12 120 0.11 0.11 0.11 0.11 4.4 Stopping Sight Distance Related 4.4.1 Stopping Distance The theoretical stopping distance is derived from the equation: SD = R T V / 3.6 + V 2 / (254 d) (5) where: SD = stopping sight distance (m) R T = perception-reaction time (s) V = initial speed of the vehicle (km/h) d = coefficient of longitudinal deceleration. This stopping distance provides the stopping sight distance standard, which is the line-of-sight distance that must be provided so that a driver can see a hazard and stop before colliding with it. Table 5 gives values of truck stopping distance derived from the semi-trailer longitudinal deceleration rates given in Table 3. The current car-based stopping distances are included for comparison. 12

Table 5. Semi-Trailer Design Stopping Sight Distances for Perception-Reaction Times of 2.5s and 2.0 s and the Corresponding Car-Based Values. Design Speed Truck-Based Car-Based (km/h) RT = 2.5 s RT = 2.0 s RT = 2.5 s RT = 2.0 s 50 68 61 54 47 60 90 82 71 63 70 115 105 91 82 80 142 131 114 103 90 172 159 140 128 100 209 195 170 157 110 258 243 205 190 4.4.2 Crest Vertical Curves As crest vertical curves serve to limit sight distance, stopping sight distance requirements impose minimum standards on this design element. Crest standards are typically expressed as a K factor relating length of curve required to the grade change: K = L / A where: K = length of curve required for a 1 % change of grade L = length of crest vertical curve (m) A = change of grade (%) When the required length of crest is greater than the sight distance, the K value is related to the sight distance standard viz: K = D 2 / C C = 200 [ h 1 + h 2 ] 2 (6) where: D = sight distance standard (m) h 1 = assumed height of eye above road (m) h 2 = assumed height of object above road (m) Note that the driver eye height assumed for trucks is greater than that assumed for cars and this will offset the effect of the difference in sight distances on the crest K value. 4.4.3 Horizontal Curves Line-of-sight obstructions on the inside of horizontal curves restrict sight distance to a value dependent on the curve radius and the lateral offset of the obstruction. Hence, sight distance requirements define standards for combinations of curve radius and the lateral offset to line-of-sight obstructions as given by: O = R [1 cos(d/2r)] (7) where: O = lateral offset to line-of-sight obstructions (m) R = curve radius (m) D = sight distance (m) For high vertical obstructions, the higher driver eye height for trucks provides no compensation for the increased stopping sight distance requirements which must be accommodated through the curve radius and offset standards. However, for low obstructions, or for non-vertical obstructions such as cut batters, the increased eye height may provide some compensation for the increased stopping sight distance. 13

5 TRUCK-BASED DESIGN VALUES 5.1 Grades Table 6 gives truck-based general maximum grades based on three design criteria: ability to maintain the design speed and ability to maintain speeds 10 and 20 km/h less than the design speed. These are given for design speeds up to 100 km/h as it is impractical to design grades for trucks operating at higher speeds. Table 7 gives length limits on grades greater than the general maximum. These are the lengths of grade at which the speed of the design truck, which enters the grade at the design speed, will have decreased by the specified speed decrement. Table 6: Truck-Based General Maximum Grade Based on Three Speed Criteria Design Speed km/h Max. Grade for Given Speed Decrement (%) 0 km/h 10 km/h 20 km/h 50 3.3 4.5 6.4 60 2.5 3.3 4.5 70 1.9 2.5 3.3 80 1.4 1.9 2.5 90 1.0 1.4 1.9 100 0.6 1.0 1.4 Table 7: Length Limits (m) of Grades Greater Than General Maximum for Specified Speed Decrements Design Speed 100 km/h 80 km/h 60 km/h Grade (%) 10 km/h 20 km/h 10 km/h 20 km/h 10 km/h 20 km/h 2 610 1,450 Unlimited 3 340 700 440 1,080 4 230 470 250 520 400 5 180 350 190 350 200 540 6 130 260 140 300 8 Not appropriate 80 170 It should be noted that the analysis carried out did not give consideration to the potential road safety problems associated with long, steep downgrades and particular attention should be given to the performance of trucks under these circumstances. 5.2 Acceleration Lanes As discussed in Section 4.2, unless combined with a downgrade, it is seldom practical to provide an acceleration lane of sufficient length for trucks to accelerate to design speeds for the through lanes. Acceleration lane lengths have been derived for three design criteria: truck accelerates to through lane truck speeds; and truck accelerates to speeds of 10 and 20 km/h less than the through speed. Table 8 gives acceleration lane lengths for these three criteria for through lane truck design speeds of 100 and 80 km/h and for level grade and various downgrades. 14

Table 8: Acceleration lane lengths (m) for trucks to accelerate from rest to a specified decrement below the through lane truck design speed. Design Speed 100 km/h 80 km/h Downgrade (%) 0 km/h 10 km/h 20 km/h 0 km/h 10 km/h 20 km/h 0 2,400 1,500 910 910 550 320 1 1,400 940 640 640 410 250 2 970 700 500 500 330 210 3 760 560 400 400 280 180 5.3 Minimum Horizontal Curve Radius Table 9a gives the minimum curve radii derived from the proposed truck side friction factor values in Table 4. The desirable minimum and absolute minimum radii correspond to the desirable maximum and absolute maximum side friction factor values respectively. For the purpose of comparison, Table 9b shows the corresponding values for car-based standards as given in the draft Austroads Guide. Design Speed (km/h) Table 9a. Minimum Curve Radii for Truck-Based SFF Criterion Desirable Minimum Radius (m) Absolute Minimum Radius (m) e = 5 % e = 7 % e = 10 % * e = 5 % e = 7 % e = 10 % * 50 76 70 64 67 62 57 60 126 116 103 98 92 83 70 200 181 159 136 127 116 80 288 258 224 199 184 166 90 410 363 310 325 295 259 100 110 As for cars 120 Table 9b. Minimum Curve Radii for Car-Based SFF Criterion Design Speed (km/h) Desirable Minimum Radius (m) Absolute Minimum Radius (m) e = 5 % e = 7 % e = 10 % * e = 5 % e = 7 % e = 10 % * 50 56 53 49 49 47 44 60 98 91 83 75 71 66 70 161 148 133 107 102 94 80 240 219 194 163 153 140 90 354 319 277 255 236 213 100 463 414 358 375 342 303 110 560 501 433 560 501 433 120 709 630 540 709 630 540 * Note: 10% superelevation is only used under extreme circumstances, usually associated with low speed alignments in confined environments. Reference should be made to the relevant Austroads Design Guide before using e values above 7%. 15

5.4 Standards Based on Stopping Sight Distance 5.4.1 Crest Vertical Curves Table 10 gives the K values derived from the truck driver eye height of 2.4 m and the Table 5 truck stopping sight distance standards. In all cases, the Table 10 K values are less than those derived for cars, indicating that the poorer braking performance of trucks relative to cars is more than offset by the greater driver eye height. Table 10. Crest K Values for Semi-Trailer Stopping Sight Distance Design Speed (km/h) Eye height = 2.4m Object height = 0.2m Crest K Value Eye height = 2.4m Object height = 0.0m RT = 2.5 s RT = 2.0 s RT = 2.5 s RT = 2.0 s 50 5.9 4.7 9.7 7.9 60 10.2 8.4 16.9 13.9 70 16.5 13.8 27 23 80 25 21 42 36 90 37 32 61 53 100 55 48 91 79 110 84 74 139 123 5.4.2 Sight Distance on Horizontal Curves Figure 2 gives relationships between lateral offset and curve radius which provide the Table 5 stopping sight distance standards for semi-trailers based on a 2.5 s perception-reaction time. For example, for 100 km/h design speed and an 800 m curve radius, a 6.8 m lateral offset from the lane centre to line of sight of sight obstructions on the inside of the curve is required to provide truck stopping sight distance. 16

Offset (m) 15 13 11 9 7 5 50 60 70 80 Design Speed 3 0 200 400 600 800 Radius (m) 15 13 Offset (m) 11 9 7 5 90 100 110 120 Design Speed 3 0 500 1000 1500 2000 2500 3000 Radius (m) Fig. 2. Lateral offsets to line-of-sight obstructions providing semi-trailer stopping sight distance standards. Note: The lateral offsets shown correspond to an assumed eye position in the centre of the lane. More accurate estimates can be obtained for a truck positioned appropriately within the lane by subtracting 0.3 m for a curve to the left and adding 0.55 m for a curve to the right. 17

6 WHEN TO APPLY TRUCK-BASED STANDARDS 6.1 Benefit-Cost Analyses Benefit-cost analysis was applied to a number of examples to estimate the truck volumes at which the adoption of truck-based standards would be economically justified. The budget and time constraints of the project necessitated a simplified and abstracted approach for these examples, with individual design elements considered in isolation. Consequently, the truck volumes for economically justified truck-based standards that follow should be regarded as indicative only. The road provision costs considered in the examples comprised earthworks and pavements only. Costs associated with structures or additional right-of-way where required were not considered. 6.2 Terrain Terrain has a significant effect on the cost of providing the more demanding truck-based standards, and hence on the truck volumes at which such standards are economically justified. Quantitative definitions of terrain have been employed earlier for road needs studies. These definitions are based on the standards of existing roads, particularly gradients, which reflect the balance between road provision costs and service to users contained in the design standards and practices at the time the road was constructed. They do not provide a quantification of terrain suitable for the consideration of alternative design standards for a new road. The definitions employed in the draft Austroads guidelines are of a more qualitative nature and relate to extended lengths of road alignment. They are unsuitable for quantifying the effects of terrain on the cost of standards for individual elements. For the purpose of the current exercise, terrain was quantified in terms of natural ground slope, with the classification given in Table 11. Table 11. Terrain Classification Based on Natural Groundslope Employed in the Benefit-Cost Analyses. Classification Typical Natural Slopes (m/m) Level < 0.03 Easy 0.05 0.10 Moderate 0.15 0.20 Difficult > 0.25 With the practice of providing above minimum standards where practical, roads designed to nominal carbased standards in level terrain will often satisfy truck-based requirements. At the other extreme, in difficult terrain truck-based standards are unlikely to be economically justified because of the high costs associated with any increase in standards. 6.3 Curve Radius and Horizontal Sight Distance The standards for horizontal curve radius and the lateral clearance to provide horizontal stopping sight distance standards are strongly related and should be considered together. Indicative truck volumes at which truck-based standards for these elements are justified are given in Table 12. The values for design speeds of 90 km/h or less apply to the provision of both horizontal curve radius and stopping sight distance. For higher design speeds, truck-based and car-based curve radius standards are the same, so that the truck volumes apply to the provision of truck-based stopping sight distance only. 18