Title: TRAFFIC LOADING
1TRAFFIC LOADING ESTIMATED FROM COUNTS
M Slavik J Bosman
2Collect
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Collect New Data
Organise
Make Accessible
Analyse
Apply
Retrieve
Knowledge
2
3- 70 freight by road
- 4 p.a. growth
- Information on traffic loading
- needed for
- pavement design
- road maintenance
- law enforcement
- statistics, patterns, trends
4- SOURCES OF TRAFFIC LOADING
- INFORMATION
- Inductive-loop counters
- Weigh-In-Motion (WIM)
- Weighbridges
5- ASPECTS OF TRAFFIC LOADING
- Magnitude
- Average Daily Traffic (ADT)
- Average Daily Truck Traffic (ADTT)
- Composition
- LV, HV HV - short, medium, long,
- buses, vehicle / axle configuration
- Axle loads
- Axle-load distribution
- ESAL (E80)
6(No Transcript)
7(No Transcript)
8(No Transcript)
9(No Transcript)
10(No Transcript)
11(No Transcript)
12(No Transcript)
13(No Transcript)
14(No Transcript)
15(No Transcript)
16(No Transcript)
17(No Transcript)
18- E80 FROM INDUCTIVE LOOPS
- Bosman 1988
- 4 classes of road, by of 2-ax HV
- Typical (default) axle-load distributions
- Bosman 2004
- Simplified to 3 classes
- Slavik Bosman in 2006
- Parameters measurable by loops
- HV - Short, Medium, Long, vs E80/HV
- Influence of law enforcement
- 3 steps
19- STEP 1 RELATION WITH E80/HV
- Loop measured attributes
- heavy vehicles (HV)
- short HV
- mediun HV
- long HV
20- STEP 2 76 WIM STATIONS, 2005
- Data validated
- Reprocessed
- Long trucks determined
- E80/HV evaluated
21- STEP 3 LAW ENFORCEMENT
- INTENSITY
- Strong permanent presence
- Medium ad-hoc, blitzes
- Weak occasional non-existent
- STEPS 1 2 3 GRAPH FIG.1
22(No Transcript)
23STEP 4 TRAFFIC LOADING MODELS Type LT class
Law Enf. 1 Below 35
Any 2 35 55 Weak 3 35
55 Strong 4 Over 55 Weak
5 Over 55 Strong
24STEP 5 FIVE MODEL STATIONS Type LT
LE Model 1 Below 35 Any
N12 Kliprivier 2 35 55 Weak
N2 Winkelspruit 3 35 55 Strong
N4 Komati 4 Over 55 Weak N3
Hidcote 5 Over 55 Strong N3
Heidelberg
25TABLE 2. Traffic and Sample Sizes at the Five
Traffic-loading Model Stations
26TABLE 3. Key Figures of the Five
Traffic-loading Types
271 - N12 KLIPRIVIER
282 - N2 WINKELSPRUIT
293 - N4 KOMATIPOORT
304 - N3 HIDCOTE
315 - N3 HEIDELBERG
32(No Transcript)
33(No Transcript)
34(No Transcript)
35- WHY?
- Weak relationship between LT and E80/HV
- Imprecision of WIM due to
- - Calibration problems
- - Deteriorating pavement
- - Hardware and software defects
- No cheap substitute for good WIM measurements
- NEXT?
- Strict WIM quality control (European Standard)
- Uniform data validation procedures
- Uniform tender requirements
- Re-appraise situation in 2-3 years time
36- CONCLUSION
- The relationship between the percentage of long
trucks and E80/HV is not very good. R-square
varies from 0,16 to 0,66. - The trend lines, however, indicate that
- the E80/HV is lower with higher law
enforcement, and - the E80/HV is higher with a higher percentage
of long trucks. - It is thus recommended that, in the absence of
better traffic loading data - the E80/HV values in Table 3 of the paper, and
- the axle distributions in Appendix A of the
paper - be used by designers and practitioners in the
meantime.
37- ACKNOWLEDGEMENTS
- The authors wish to express their gratitude to
- NTRV (Northern Toll Road Venture, the N1 Toll
Road Concessionaire) -
- N3TC (N3 Toll Concession, the N3 Toll Road
Concessionaire), - TRAC (Trans African Concessions, the N4 Toll
Road Concessionaire), -
- Bakwena (the N4 Platinum Toll Road
Concessionaire), and - SANRAL (South African National Roads Agency
Limited) - for the traffic data and information made
available.
38TRAFFIC LOADING ESTIMATED FROM COUNTS
M Slavik J Bosman
39(No Transcript)
40(No Transcript)
41(No Transcript)
42(No Transcript)
43(No Transcript)
44(No Transcript)
45- FACTS
- Bad news
- Too dispersed - bottom of a birds cage
- Good news
- Red line above green line
- All lines rising from left to right
- Blue line joins green on left, red on right
46TABLE 1. Five Traffic-loading Model WIM
Stations