Estimating average daily traffic using alternative method for single carriageway road in Southern Region Malaysia

• Main Authors: Nor, N. S. M., Puan, O. C., Mashros, N., Ibrahim, M. K. B.
• Format: Article
• Language: English
• Published: Asian Research Publishing Network 2016
• Subjects: TA Engineering (General). Civil engineering (General)
• Online Access: http://eprints.utm.my/id/eprint/74470/1/NordianaMashros2016_EstimatingAverageDailyTrafficusingAlternative.pdf; http://eprints.utm.my/id/eprint/74470/; https://www.scopus.com/inward/record.uri?eid=2-s2.0-85009204418&partnerID=40&md5=16ee6c0f102365280a06fbe4482773a2
• Institution: Universiti Teknologi Malaysia

Average Annual Daily Traffic (AADT) and Average Daily Traffic (ADT) are two parameters that are commonly required by traffic engineers and road designers to design and analyse the traffic operational performance of a road segment. In Malaysia, ADT is normally used to forecast the volume of traffic in the design year as well as to design the pavement thickness. Basically, ADT can be generated using expansion factor estimates from Peak Hour Volume (PHV). Current practice in Malaysia uses an expansion of 10% to estimate ADT from PHV. This paper discusses the results of a study carried out to establish a model for estimating ADT using PHV for single carriageway road. The 24-hours traffic data were collected at 9 sites in the districts of Johor, Malaysia for the period of 14 days. The 7-days data were used to establish the model and the other 7-days data were used to validate the model. For validation purposes, the absolute percent error (APE) for each estimate of ADT obtained from the model was calculated and compared with observed ADT. The statistical test at 95% confidence level was conducted to determine the significance difference between the ADT from actual data and the estimate ADT from model. The result shows that a power-formed trend line (y=axb) suits to the observed data with the coefficient of determination of about 0.90. Validation result shows that the ADT for the model has lesser APE compared with the ADT estimated using the factoring approach. A comparison of both estimated and actual ADT values using t-Test shows that there is no significant difference between the estimated ADT using models and the actual ADT. However, the ADT estimated using the expansion factor of 10% shows the vice versa. Such a finding implies that the model obtained from this study predicts ADT accurately than the current practice.