TRANSPORT RESEARCH LABORATORY TITLE by Preliminary analysis of exclusive motorcycle lanes along the federal highway F02, Shah Alam, Malaysia Radin Umar R.S., Murray G Mackay and Brian L Hills Overseas Centre Transport Research Laboratory Crowthorne Berkshire United Kingdom IA r
_. -- UMAR, R S, M G MACKAY and B L HILLS (1995). Preliminary analysis of exclusive motorcycle lanes along the federal highway F02, Shah Alam, Malaysia. IA TSS Research vol. 1.9 No 2. 1995. PA 3136/96 PRELIMINARY ANALYSIS OF EXCLUSIVE MOTORCYCLE LANES ALONG THE FEDERAL HIGHWAY F02 R MR ..MCABL IL PRELIMINARY ANALYSIS OF EXCLUSIVE MOTORCYCLE LANES ALONG THE FEDERAL HIGHWAY F02, SHAH ALAM, MALAYSIA Radin UMAR R.S. Assoc. Professor, Accident Research Unit, Faculty of Engineering, Universiti Pertanian Selangor, Malaysia Murray G. MACKAY Professor, Accident Research Centre, Birmingham University, England Brian L. HILLS Safety Manager, Overseas Centre, Transport Research Laboratory, England (Received July 10, 1995) One way to tackle the problem of motorcycle accidents between junctions (links) is to segregate motorcycles from other traffic using an exclusive motorcycle lane. This paper presents a preliminary analysis on the impact of the motorcycle lane in reducing link motorcycle accidents along Federal Highway F02, Malaysia. The study has shown that there has been a significant reduction (p<0.05) in motorcycle accidents following the introduction of the lane. The overall reduction of all motorcycle accidents along the 14 kilometre stretch is approximately 25%. When stratitied motorcycle accidents are considered, the reduction of motorcycle accidents excluding the those that occurred on the main carriageway, is approximately 34%. Both reductions are highly significant at 5% (p<0.05) and 0.1 % (p<0.005) for all motorcycle accidents and stratified motorcycle accidents respectively. These findings support the notion that segregating motorcycles from other traffic reduces the accident exposure and improves significantly the safety of motorcyclists. Key words: Motorcycle accidents, Mixed flow, Traffic segregation, Motorcycle lanes 1. INTRODUCTION On roads with high level of traffic, conflicts between vehicles may be created when heavy commercial vehicles and fast moving cars have to share the same roadway fa- cility with the less protected and slower vehicles such as motorcycles, mopeds and bicycles. On these road sections, separation of motorcycle flows from other traffic will not only improve link accidents with motorcycle but can also improve the traffic flow especially when motorcycle traf- fic is heavy. The concept of segregating vulnerable road users from other traffic is not new in the field of safety management. In many industrialised cities, traffic segregation, whereby pedes- trians and bicyclists are segregated from motorised traf- fi c 12'3,4, has long been recognised and forms part of the planning framework in the transport planning process. In de- veloping countries, a number of studies 5'6have also been car- ried out on the potential use of road marking installations for mixed flow conditions. Unfortunately, there has been little work and experiences so far on the use of exclusive tracks for motorcycles. Lack of commitment to plan safety strate- gies and financial constraints usually combine and result in shelving these measures. Perhaps, Malaysia might be the first country to implement this measure to promote safety for motorcyclists. 2. USE OF MOTORCYCLE LANES IN MALAYSIA The notion of segregating motorcycles from other traf- fic by the use of a motorcycle lane is not new to Malaysia. In the early seventies, the first motorcycle lane was con- structed along the Federal Highway Route 2, one of the busi- est urban expressways in the country, under a World Bank project. The track consists of an exclusive 16 kilometer mo- torcycle track connecting the city of Kuala Lumpur and the Subang International Airport. For safety and accessibility of motorcyclists, grade-separated interchanges are used to con- nect the track with the main carriageway. Unfortunately, no study has so far been conducted to analyze the safety and eco- nomic benefit of the scheme. In early 1992, an extension of the track was carried out. This extension was part of the improvement programme to the existing two-lane expressway connecting the Subang In- ternational1 Airport to the towns of Shah Adamn and Klang. The project was constructed and managed by Projek Lebuhraya Utara Selatan (PLUS), under one of the government's privatisation schemes. In November 1993, major sections of the lane were completed ahead of schedule and it was opened for use by the public. This paper presents a preliminary analy- sis on the impact of the motorcycle lane on motorcycle acci- dents along Federal Route F02, withfin the city of Shah Adam, Malaysia. IA TSS Research Vol. 19 No.2. 1,995 * 93 R. UMAR, M.G. MACKAY, 131. HILLS CURRENT TRANSPORTATION ISSUES FROM ASIA 3. METHODOLOGY This study was based primarily on a specially created police accident form POL27(Pin 1/9 1) and the 24 hour po- lice accident reports. The former was designed for easier completion and fully compatible with a customised version 7 of the TRL's Microcomputer Accident Analysis Package, MAAP 8.Accident data were extracted from the four-year pilot project data 9"10obtained in the districts of Seremban and Shah Alam since January 199 1. For each type of motorcycle accident, a detailed col- lision mechanism that lead to the accident was reclassified and recoded based on sketch diagrams and written police descriptions available on the last page of the police form. A total of 193 collision mechanisms divided into 37 col- liding groups were identified" and summarised in Figure 1. The colliding group was classified according to the road geometry, vehicle manoeurvres and the right-of-way (ROW) with the notations S to represent all straight direc- tions on ROW, T for turning on the ROW and X for cross- ing manoeuvres on losing ROW. Collision types were also identified by the impact directions with notations F, B and D for the front, rear and side collisions respectively. These recoded records were then manually updated into MAAP. Analysis of the data was carried out using the cross- tabulation facilities available in MAAP, the time series cu- mulative plot' 2of monthly records and the traditional Chi-squared "before" and "after" analysis' 3"14.In this pre- liminary analysis, only the total motorcycle accidents were used. The detailed colliding mechanisms described in Fig- ure 1, however, will be extensively used in the detailed and long-term analysis of the lane. 4. 'BASIC FEATURES OF THE MOTORCYCLE TRACK Figure 2 shows a typical view of the extended track along the Federal Highway Route 2, Shah Alam. The ex- tended track consists of a 14 kilometre exclusive track run- ning alongside with the main three-lane urban expressway. The basic features of the motorcycle track are shown in Table 1. 94 * /A TSS Research VoL.19 No. 2. 1995 Table 1 Basic features of motorcycle lane along Federal Highway Route 2 Motorcycle Lane Features Design Parameter Track length (Extension) 14 Kilometer Track Width 2.5- 3.5 meter Verge 1 -2 meter Distant from Main Varies with maximum Carriageway 3 meter Access Control Full Guardrail Type Single face Wearing Course Formulation 50mm Asphatic Concrete Road Base Formulation 150mm Wet Mix Macadam Sub-Base Formulation 250mm CBR>30% Interchange TypeGrade-Separated (Courtesy from PLUS) 5. IMPACT OF MOTORCYCLE LANES ON ACCIDENTS 5.1 Time series plot of motorcycle accidents The pattern of motorcycle accidents along the Fed- eral Highway Route F02, 6 months following the introduc- tion of the motorcycle lane is shown in Table 2. The cumulative plot of motorcycle accidents along the route and all accidents in Shah Alarn is shown in Figure 3. In this analysis, all accidents in the city of Shah Alam were used as the control data. The rationale of chosing this as the con- trol instead of all accidents along the whole route are (a) the main carriageway was also upgraded from the two-lane to three-lane expressway (b) the number of accidents ex- ceeds (about 18 times higher) the number of motorcycle accidents. From the above illustrations, it can be seen that the number, of motorcycle accidents dropped markedly imme- diately following the opening of the motorcycle lane.The overall drop of motorcyce accidents along the route was approximately 25%. This drop can also be seen graphically from the clear downward separation of cumulative num- ber of motorcycle accidents with respect to the predicted cumulative mean of accidents. In contrast to the reduction of motorcycle accidents, the number of all accidents in the control area remains steady at an average rate of 235 accidents per month. The I ' .
' j : 7 ' ..I'


. .' . .. 1 . . 1 .- .' --- a
1 ' . .
:. 1 . ---- --- . .. .. .m .1 . 1 . PRELIMINARY ANALYSIS OF EXCLUSIVE MOTORCYCLE LANES ALONG THE FEDERAL HIGHWAY F02 A MR ..MCABL IL Colliding Machanism Motorcycle Straight Pedestrian Cross Motorcycle Turn Pedestrian Cross Motorcycle Straight Pedestrian Obscured Motorcycle Straight Pedestrian Middle of Road Motorcycle Straight Pedestrian Walk Along Road Motorcycle Straight Pedestrian Walk Opposite Direction Motorcycle Straight and Hit Objects Motorcycle Straight and Hit Animals Motorcycle Straight and Hit From Back Motorcycle Turn and Hit From Back Motorcycle Cross and Hit From Back Motorcycle Cross and Hit From Front Motorcycle Cross and Hit From Side Motorcycle Turn and Hit From Side Motorcycle Turn and Hit From Front Motorcycle Straight Other Vehicles Cross Side Collision Motorcycle Straight Other Vehicles Cross Front Collision Motorcycle Straight Other Vehicle Cross Rear-End Collision C r ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 Examples Notation I Colliding Machanism Motorcycle Backing Other Turn Rear-End Collision Motorcycle Straight Other Straight Side-Swipe Collision Motorcycle Straight Other Straight Head-On Cillision Motorcycle Backing Other Straight Rear-End Collision Motorcycle Park/Stop Other Straight Rear-End Collision Motorcycle Straight Other Park/Stop Rear-End Collision Motorcycle Straight Other Park Side Collision Motorcycle Straight Other Reverse Front Collision Motorcycle Yaw/Fall and Hit Pedestrian Motorcycle Yaw/Fall Ott the Road Motorcycle Yaw/Fall On the Road Motorcycle Queue at Stop Line Other Turn Hit From Front Motorcycle Straight Other Yaw Head-On Collision Motorcycle Yaw/Fall Other Straight Head-On collision Motorcycle Straight Straight Other Yaw Rear-End Collision Motorcycle Yaw/Fall Other Straight Rear-End Collision Mototcycle Park Other Turn Rear-End Collision MBOTB MSOSD MSOSF MBOSB MEOSEB MSOEB MSOED MSORF MCYHP MYOFF MYONR MOOTF MSOYF MYOSF MSOYB MYOSB MEOTB --- Mc 55. 717 Ill --- ,Wm.- _=_11 _T 7
- tla. -E

IT- ii
c
:
rr-
,g
I .i* _____________________________________________________________________ L Examples -T 7 -2 M -E:a -,- e- 2 K ,A -P..- 4 Q D~_ lc D' L Fig. 1 Detailed colliding mechanisms involving motorcycles in Malaysia 1ATSS Research Vol. 19 No.2, 1995 * 95 Notation MVCSpx MCTIPX MSPO13 MSPMR MSIPWA MSPWO MSH0,1 MCSHA MCSHB MCTHB MCXHB MCXHF IVCXHD MCTHD MVCTHF MSOXID MSOXF MSOXB t 1 R. LIMAR, M.G. MACKAY, 13.1---. HILLS -pf!2
, 70yr- CURRENT TRANSPORTATION ISSUES FROM ASIA 01E3 .40 2 E4 3 May July Sept Nov JuuSA Mar May Jul Apr Jun Aug Oct Dec3 Fula Apt Jun Aug94 Month at tha Y-a (1 95311994) 500 400 300 5 1 0 4 100 0 Fig.2 Typical view of motorcycle track along Federal Highway Route 2, Malaysia Fig. 3 Cumulative plot of motorcycle accidents along F02 and all accidents in Shah Alam Table 2 Motorcycle accidents along Federal Route F02 Shah Alamn Before Period ( March 1993- November 1993) Accident Type Mar93 Apr May June July Aug Sept Oct Nov Motorcycle Accidents along F02 1Ž> 7 16 18 9 1 1 16 16 1 7 Cumulative M/C Accidents F02 1 2 19 35 53 62 73 89 105 122 Cumulative Mean M/C Accidents 1 2 26 39 53 66 80 93 107 120 All Accidents (Control) 236 256 237 230 230 225 232 267 205 Cumulative Control 236 492 729 959 1189 1414 1646 1913 2118 Cumulative Mean Control 236 471 707 942 1177 1413 1648 1883 2119 After Period (December 1993 -August 1994) ____________ Accident Type Dec93 Jan94 Feb Mar Apr May June July Aug94 Motorcycle Accidents along F02 2 13 8 10 16 12 9 13 10 Cumulative M/C Accidents F02 124 137 145 155 171 183 191 204 214 Cumulative Mean M/C Accidents 134 148 161 174 188 202 215 229 242 All Accidents (Control) 200 237 204 231 266 265 252 264 283 Cumulative Control 2318 2555 2759 2990 3256 352 ~3773 4037 4320 Cumulative Mean Control 2354 2589 2824 3060 3295 3530 3766 14001 14236 number of accidents in the control area in the before pe- riod was 2118 while the after period was 2202. To ascertain that the reduction is statistically significant, a more detailed anaysis is required. This can be achieved us- ing the standard Chi-squared before and after study. 5.2 The Chi-Squared (X2) "before" and "6after" analysis The figures in Table 3 show a cross tabulation of ag- gregated all motorcycle accidents along F02 "before" and "after" the opening of the motorcycle lane. In this analy- A', PRELIMINARY ANALYSIS OF EXCLUSIVE MOTORCYCLE LANES ALONG THE FEDERAL HIGHWAY F02R.MAG.ACYBLHIS Table 3 Short -term analysis on motorcycle accidents along Federal Highway Route F02 Accident Type Before Period After period Reduction x2 Probability _____________________________ (March -Nov93) (Dec93 -August94) (%) ____________ All Motorcycle Accidents Along F02 122 92 25 5.19 p<0.05 Stratified Motorcycle Accidents (F02) -122 81 34 9.51 p<0.O01 All Accidents in Shah Alamn (Control) 2118 2201 __________ _____ _________ sis, the motorcycle accidents in the "after" period were fur- ther classified into two categories: (i) all accidents along the route, and (ii) all accidents along the lane. The rationale for this stratification is to separate all motorcycle accidents along the main carriagew ay and the accidents along the cycle track. This will allow for a more comprehensive assessment of the: (i) actual impact of the lane, (ii) compliance of motorcyclists to use the cycle lane, (iii) the possible effect of incomplete track due to site constraints at kilomneter 18. It is worth mentioning that a small section of the cycle lane at kilometre 18 was still under construction af- ter the official opening of the lane. This apparent delay was not due to any management and construction problems by PLUS but rather due to the fact that the whole construc- tion was well ahead of the original schedule. Since it was wasteful to leave the lane unused and to wait for the origi- nal schedule, the management had decided to open the lane to the public at the earliest opportunity. During this study period, motorcyclists however had to join the main express- way for a short distance before rejoining back to their cycle track. Based on this "before" and "after" analysis, it can be stated that the introduction of a motorcycle lane has sig- nificantly reduced motorcycle accidents along the route. The computed X2value for one degree of freedom is 5.19 and the reduction is significant at 95 percent confidence level (p<0.05). When stratified data is considered, the im- pact of the motorcycle lane in reducing motorcycle acci- dents is more distinctive with the X2value of 9.51 (p<0.005). 6. DISCUSSION The above study deals only with the short-term im- pact on the use of motorcycle lane along Federal Route 2, Shah Alam, Malaysia. This preliminary analysis has shown the introduction of the lane has significantly reduced mo- torcycle accidents along the route. No direct comparison can be made on the benefits of the motorcycle lanes, since no other research has been car- ried out or reported before. However, the success of this traffic segregation scheme is in line with the concept of seg- regating bicycles from other traffic in many developed countries. In Japan', it was reported that vehicle to bicycle accidents decreased by 15.4% following the introduction of bicycle-pedestrian path. This study was based on a be- fore and after comparison of 96 locations in Japan where the path had been installed between January 1986 and June 1987. In another study on the effect of the Western ap- proach cycle route in Southampton 3, it was reported that although there was no change in the total number of about 12 bicycle accidents per year on the route, cycle flow in- creased by 28% when flows declined 14% elsewhere in the city. Thus, it was interpreted that the initiative reduced the cycle accident rate along the route. Although the exact explanation of the reduction of motorcycle accidents following the lane construction is yet to be examined, a number of factors can be put forward to the success of the scheme. Segregating slow moving mo- torcycles from fast moving traffic, particularly the low cc motorcycles, would eliminate traffic conflicts and speed differentials between motorcycles and faster vehicles. This idea is in agreement with the fact that approximately 99% of the rear-end accidents with motorcycles between 1991 and 1993 in this pilot areas involved lower cc, slow mo- torcycles. This concept is also in agreement with Munden' 5 who established the U- shaped variation of accident rate IA TSS Research Vol. 19 No.2, 1995 * 97 R. UMAR, M.G. MACKAY, 131. HILLS - .. . ... CURRENT TRANSPORTATION ISSUES FROM ASIA and relative speed; the fastest and slowest drivers have higher accident rates than the average speed drivers. Thus segregating motorcycles from other traffic will eliminate these speed differentials, provided that they stay away from the main carriageway. In this research, it has also been shown that when data are stratified according to the exact location of the acci- dents, whether on the track or on the main carriageway, the reduction in motorcycle accidents is more significant in the case of stratified data. This implies that there is still a sizeable number of motorcyclists who stayed on the main carriageway instead of riding on the cycle lane. This might be the result of (i) the higher cc large bikes which were given the option to use either route of the expressway and (ii) the enforced joining of the main traffic at kilometre 18 of the carriageway. However, no research has been carried out to substantiate this hypothesis. The change in riding environment is one of the im- portant factors in influencing the safety of motorcycles. This is because, riding a motorcycle differs fundamentally from driving a car in terms of its stability. Adverse weather conditions, poor road surfaces such as pot holes, oily and sandy roads' 6and aerodynamic disturbances from larger vehicles' 7can easily cause a motorcycle to become un- stable. Thus creating a safer riding environment by means of exclusive lanes would contribute to the lower number of these types of motorcycle accidents. 7. CONCLUSIONS This preliminary analysis suggests that traffic segre- gation by means of exclusive motorcycle lanes may prove to be one of the best ways to achieve the desired safety ob- jective particularly along routes with a high population of motorcycles. This is because it reduces the number of mo- torcycle accidents dramatically following the intervention. The overall short-term reduction was found to be 25% and significant at the 5% (p<0.05) level. If accidents along the main carriageway are excluded, the reduction was 34% and significant at the 0.5% (p<0.005) level. Besides solving the problems of speed differentials and mixed flow problems of the incompatible traffic, this measure also creates bet- ter driving environments for both motorcyclists and other road users. 98 * IATSS Research Vol. 19 No.2, 1995 REFERENCES 1 .Organisation for Economic Co-operation and Development, Safety of two-wheellers, Paris March (1978). 2. Noordzij P.C. Cycling in the Dark:An analysis of fatal bicycle accidents, "Journal of Safety Research", Vol 8. No 2. The Netherlands (1976). 3. Morgan J., Cycling in Safety?. Proc. Safety 91 Seminar. Transport and Road Research Laboratory, Crowt homne, England (1991). 4. Matsumura T, Seo T., Umezawa M, and Okutani T. Road Structure and Traffic Safety Facilities in Japan, Proc. Int. Conference on Asian Road Safety CARS'93, Kuala Lumpur (1 993). 5. Chari S.R. and Nath K.M. Accident Causative Factors under Mixed Traffic Conditions- A case Study of Hyderabad City. Proc. International Conference of Traffic Safety, ICOTS91, New Delhi India (1991). 6. Chen C. Study on Road Marking Installation of Mixed Flow, Proc. 4th Conference Road Engineering Association of Asia and Australasia, Kuala Lumpur (1983). 7. Radin Umar R.S., Ahmad Rodizi M and Aminuddin A, The Malaysian accident diagnosis and treatment mode, "Pertanika Joumnal Science and Technology", Vol 1 No 1. Universiti Pertanian Malaysia, Serdang, Selangor (1993).(ln Malay). 8. Hills B. Elliot G and Clark D, MAA1P S User Guide v5, Overseas Centre, Transport Research Laboratory, Crowthorne, United Kingdom (1994). 9. Radin Umar R.S and Aminuddin A , A Pilot Project on the Accident Diagnostic System in Malaysia, Research Report No 1, JK3P, National Road Safety Council, Malaysia(11992)(ln Malay). 10. Radin Umar R.S and Aminuddin A. Accident Diagnosis System. 32nd Annual Seminar, National Road Safety Council Malaysia. Kuching, Sarawak, Malaysia (1991 )(ln Malay). 1 1. Radin Umar R.S. Mackay G.M. and Hills B.L..- Preliminary analysis of motorcycle accidents: short-term impacts of the running headlights campaign and regulation,"Journal of Traffic Medicine', United Kingdom (1995). 12. Institution of Highways and Transportation, Guidelines for Urban Safety Management. London, United Kingdom (1992) 13. Royal Society for the Prevention of Accident, Accident Investigation Manual, Canon House, The Priority Oueensway, Birmingham, United Kingdom (1 986). 14. Tanner J.C, A Problem in the combination of accident frequencies, "Biometrika", Vol 45, Part 3 and 4, United Kingdom (1 958). 15. Munden J.M. The Relation Between a Driver's Speed and his Accident Rate, Road Research Laboratory Laboratory Report LR 88. Crowthome, United Kingdom (1967). 16. Ouelett J.V. , Envoronmental Hazards in Motorcycle Accidents, 26th Annual Proceedings. American Associoation for Automobile Medicine, Ontario. Canada (1982). 17. Brosson B and Efver J, Wobbling in modern motorcycles, 'Accident Analysis and Prevention", Vol 16 NO 5/6, pp 451 -456 (1984). ' p. 1 .1 1 .1 ; .: 1 .