TRANSPORT and ROAD RESEARCH LABORATORY. Department of the Environment Department of Transport TRRL SUPPLEMENTARY REPORT 811

Transcription

1 TRANSPORT and ROAD RESEARCH LABORATORY Department of the Environment Department of Transport TRRL SUPPLEMENTARY REPORT 811 SPECIALISED BUS SERVICES IN THREE THIRD WORLD CITIES by D A C Maunder and P R Fouracre Any views expressed in this report are not necessarily those of the Department of the Environment, Department of Transport or the Overseas Development Administration Overseas Unit Transport and Road Research Laboratory Crowthorne, Berkshire 1983 ISSN

2 Ownership of the Transport Research Laboratory was transferred from the Department of Transport to a subsidiary of the Transport Research Foundation on I st April This report has been reproduced by permission of the Controller of HMSO. Extracts from the text may be reproduced, except for commercial purposes, provided the source is acknowledged.

3 CONTENTS Page Abstract Introduction High quality services at premium fares 2.1 Operational characteristics 2.2 User characteristics Incomes Journey purpose and trip frequency Reasons for choice of mode Special types of service 3.1 Operational characteristics 3.2 User characteristics Incomes Journey purpose and trip frequency Reasons for choice of mode Services to attract car users to public transport 4.1 Operational characteristics 4.2 User characteristics Services which improve the operators productivity 5.1 Operational characteristics 5.2 User characteristics Discussion Summary and conclusions Acknowledgement References CROWN COPYRIGHT 1983 Extracts from the text may be reproduced, except for commercial purposes, provided the source is acknowledged

4 SPECIALISED BUS SERVICES IN THREE THIRD WORLD CITIES ABSTRACT In recent years operators of public transport bus services in Third World cities have increased their operation of dual services. As well as ordinary stage-carriage services, specialised services (providing, for example, more speed and comfort) are operated mainly on high demand corridors or to middle-income residential areas. The report describes the operational characteristics of specialised bus services in two Indian cities, Hyderabad and Delhi, and in Bangkok, Thailand. In addition, socio-economic characteristics and trip details of passengers are examined and an assessment made of the extent to which the services meet the objectives expected of them. It is possible that dual services, such as those described, may become the normal operating system of the future in most Third World cities. A basic and inexpensive ordinary service can be operated for low income users and a more expensive service operated for middle income users who can better afford the cost of higher quality. 1. INTRODUCTION Over recent years stage carriage public transport operators in many Third World cities have increased the supply of dual bus services on some highly patronised routes and/or to certain sectors of the city. The dual service consists of an ordinary conventional bus service plus a faster, more comfortable (often seating-only) and sometimes air-conditioned service at a premium fare level. The quality services tend to be operated to middle income residential areas where there is potential demand for a high quality service or on high demand corridors linking the city centre to residential and commercial centres. Other premium fare services are operated during unsocial hours and for provision of extra facilities (like luggage space). All such services are referred to in this report as 'specialised'. Although the reasons for introducing these services in particular instances are not always apparent, there are possibly four objectives for the operation of specialised services:- (i) To provide for those passengers who are willing and able to pay premium prices regularly for a superior level of service and in so doing to maximise revenues. The limited stop and deluxe Mudrika service operated by the "Delhi Transport Corporation (DTC) in Delhi are included in this category. The Andhra Pradesh State Government sponsored Society for Employment, Promotion and Training in Hyderabad and Secunderabad (SETWlN) also operates a few conventional single deck buses supplying a limited stop service in Hyderabad. Privately operated charter buses should be included in this category and also the air conditioned service operated in Bangkok by the Bangkok Metropolitan Transport Authority (BMTA). For the purposes of this report, however, the BMTA luxury buses are discussed under a separate heading. (ii) To provide a distinct and special type of service (which would not otherwise be operated) which, though costly to provide and use, presents a cheap alternative to taxi-like services. The DTC provision of railway specials and night services comes into this category. The provision of these services is largely designed for infrequent users. (iii) To attract potential passengers who presently travel by personal vehicles such as cars and motorcycles, thus reducing traffic congestion and air pollution in cities. In addition, where countries have to import large

5 quantities of oil to meet domestic requirements, a transfer of mode from personal vehicle to public transport services helps conserve fuel. The air conditioned service operated by BMTA in Bangkok was introduced principally for these reasons. (iv) To increase the productivity of vehicles and crew. With faster, limited stop services, the output of vehicles (such as kilometres operated and turnover of passengers) can be increased and a greater contribution to running costs should be possible. Here the emphasis is on vehicle productivity rather than fare differentials (maximisation of revenue per passenger). An example of such a service would be the express services operated by the Andhra Pradesh State Road Transport Corporation (APSRTC) in Hyderabad, which are operated at a fare level which is only marginally above that of the ordinary bus service. It might be expected that bus services in any of the first three categories will appeal to specific groups of traveller, be they from higher income communities, car-owners, those wishing to make a one-off journey to the railway station, etc. Their provision is based on the assumption that the market for bus services is not homogeneous and that some users value the elements of service quality more highly than others. The fourth category of specialised service is more a means of improving the operator's efficiency. If fare differentials between the ordinary and express service are minimal then there is no reason to suppose that users of either service will have significant differences in socio-economic background. This report presents an examination of the operations of a number of specialised bus services to see how well they meet these various objectives. Surveys were undertaken in Delhi, Hyderabad and Bangkok of the services briefly introduced above. The surveys of the two Indian cities were made during the two year ( ) joint research programme between the Association of State Road Transport Undertakings (ASRTU) and the TRRL 1. The surveys of the specialised services of BMTA in Bangkok were undertaken by the Overseas Unit, TRRL, in March The report is structured so that in the following sections each of the operating objectives is presented in turn and illustrated in terms of one, or more (as appropriate) of these examples. 2. HIGH QUALITY SERVICES AT PREMIUM FARES 2.1 Operational characteristics The main example chosen for study is that of the limited stop service operated by the DTC in Delhi. Two other examples, the charter buses of Delhi and the SETWlN limited stop buses in Hyderabad, are briefly noted at the end of this section, though few detailed surveys of their operational performance were undertaken. The DTC operates four limited-stop services from various parts of the city, to two locations in New Delhi and one in Old Delhi. The service routes are operated from dense, heavily-populated, residential areas on the outskirts of the city to the commercial area of New Delhi (Super Bazar), government office area (Kendriya Terminal) and commercial area of Old Delhi (Red Fort). At the time of the study they were only operated in the peak demand period and in the peak direction and are scheduled to operate when most offices, both government J.. and commercial, open and close for business. A single service is operated in the peak direction on three of the routes served, whilst from Janakpuri, a mid-upper income area located 20 km from the city centre, two services are operated in the morning and one in the reverse direction in the evening. The vehicles used on each of these routes are used on other routes before and after the peak time limited stop service so that both crew and vehicle are fully occupied throughout the operating day. The vehicles and crews providing this additional service on high demand corridors are withdrawn from other less heavily patronised routes.

6 The type of vehicle used on the limited stop service is shown in Hate 1. It has two distinguishing features; at the front and rear are signs with the DTC logo which denote the service (these are however difficult to see) and secondly the stand plate (which is much easier for passengers to see), which identifies the major stands serviced and level of fare for the service. The stand plate is located at the entrance of the bus. A flat fare of Rel is charged which is higher than that paid by passengers using ordinary services. The ordinary service fare levels are:- 30 paise for distances of up to 4 kin; 40 paise for distances of 4 to 16 km; 50 paise for distances of km; and 75 paise for distances over 20 kin. Table 1 shows monthly operational characteristics of the four limited stop services for December TABLE 1 Operational data of DTC limited stop services for December 1979 Route Route distance (krn) Service start time Schedule Kilometres Actual Revenue (Rs) Earnings per km (paise) Saket to Super Bazar Janakpuri to Red Fort Tikri Border to Cent. Sectt Ashok Vihar to Cent. Sectt Source: DTC records December 1979 (Note: paise is equivalent to Re1, and in was equivalent to Rs 18). One of the most commonly used measures of performance in the bus industry is the Earnings per Kilometre (EPK), which is the ratio of bus revenues to bus km operated. Most routes in Table 1 show high EPK values; on only one route (Central Secretariat to Tikri Border) is the EPK considerably lower than that for the whole DTC network, which in December 1979, was 174 paise per km. (However, it must be observed that proper comparison should only be made between data for peak-time trips in the peak direction as shown below.) The limited stop service on the route linking Janakpuri to Red Fort was monitored over a six day period in July, 1979, when only two trips per day were scheduled. During this period, the number of passengers per bus 3

7 trip in the morning ranged from 61 to 78, with an average of 73. For the evening return at 17.30, the number of passengers ranged from 35 to 93 with an average of 58. The average revenue per trip at that time was, therefore, Rs 66, which is compared in Table 2 to peak revenue figures for two other long-distance routes carrying ordinary services 2. TABLE 2 Comparative earnings on limited stop and ordinary bus routes, Delhi Route Average bus load per peak trip (peak direction) Peak revenue (Rs) (peak direction) Peak EPK (paise) Peak direction Return direction Estimated load factor (peak direction) Limited stop: Janakpuri to Red Fort * 42* n.a ** * Average fare per passenger of 40 paise ** No standing permitted Routes 720 and 155 are typical of the extremes of normal and exceptional overcrowding on buses in Delhi in the peak. The EPK figures are high under these conditions. Though this does not provide conclusive evidence, it is apparent that revenues earned by limited stop services may not be exceptional and that revenue maximisation is not necessarily achieved through this policy of dual services. Though the operators of the limited stop services may not benefit greatly from the exercise, it is reasonable to assume that the additional cost that users incur is offset by extra comfort (lower load factors). The extra fare also buys a faster journey time on some links. The Janakpuri -Red Fort limited stop service links Janakpuri directly with major city locations such as Kendriya Terminal, Connaught Place, ITO and Red Fort. Table 3 compares the scheduled trip times and fares of DTC ordinary and limited stop buses linking Janakpuri to these locations. TABLE 3 DTC services operated from Janakpuri to Central area Janakpuri to Trip times (mins) Fares (paise) Ordinary Limited stop Ordinary Limited stop Kendriya Terminal Connaught Place ITO Red Fort No direct service n.a. 75 For passengers travelling to Kendriya Terminal (the major passenger flow), the limited stop service provides a much faster trip than the ordinary service at an additional fare of 25 paise. Trip times from Janakpuri to Connaught Place are similar by both services while to Red Fort the trip time by limited stop service is considerably

8 greater than that by ordinary bus; in both cases the ordinary bus is operated on a more direct route and therefore trip times are not directly comparable. Table 4 shows observed vehicle speeds for ordinary buses (on route 720 between Janakpuri and Kendriya Terminal) and the limited stop service. TABLE 4 Average trip speeds: Route Janakpuri - Kendriya Terminal Time period Direction Service Speed kph Peak towards city centre towards Janakpuri limited stop ordinary limited stop ordinary For both services the average trip speed was similar with the limited stop service only marginally faster than the ordinary service. The comparison is not entirely valid, however, as the limited stop service covers some very congested roads not covered by the ordinary service. This congestion occurs on the section from Kendriya Terminal to Red Fort. Over similar route sections like Janakpuri to Kendriya Terminal, the time saving of almost 25 per cent (Table 3) is probably a fair reflection of the improved performance of the limited stop service. Such savings in trip times are unlikely to be reflected in better vehicle utilisation, since these limited stop trips constitute only a small portion of the vehicle's total daily output. Apart from the DTC limited stop services charter bus services are provided from various parts of Delhi mainly for commuters from the suburbs of Delhi to the offices and commercial centres. Charter bus services are privately operated and chartered by residents of a colony or area to provide a service between the residence and commerical centre. Over such services are provided daily to and from the residential area of Janakpuri alone and approximately are provided daily throughout Delhi. These services are only provided in the direction of peak demand with each bus usually providing a single service trip, though some provide two trips. Services are provided between from Janakpuri and back to Janakpuri. The vehicles are used for other charter work, such as transporting school children, during the rest of the day. A monthly tariff is charged from each member of the commuter group to cover the hire charges, and is equivalent to approximately 80 paise per passenger trip. Guest passengers are carried where seating capacity is available and they pay Rel. The prolific expansion of these services in Delhi in recent years is due to the consistent operation of the service and the comfort and speed of travel. Vehicle speeds tend to be similar to or slightly more than those of limited stop services, with fewer pickup points en route. No surveys were undertaken to establish operational data and trip revenues on Delhi charter buses. Since they are privately operated vehicles it can reasonably be assumed that both revenues and profits earned on these trips are good. Earnings are likely to be as high as those of the DTC limited stop services and load factors at or close to unity. In Hyderabad, limited stop services were introduced by SETWlN during October The service employs conventional 50-seat, single-deck vehicles (see Hate 2) on a route network which is controlled by the monopoly stage carrier, APSRTC. The service is seating-only and the fares charged are higher than those charged by APSRTC for ordinary services. The service is not confined to peak-hour operations being provided throughout 5

9 the operational day. Table 5 contains some comparative revenue data obtained from limited surveys of two parallel routes, one operated by SETWlN (limited stop) and one by APSRTC (ordinary). TABLE 5 Comparative earnings on limited stop and ordinary bus routes, Hyderabad Route Average bus load per trip Average fare per passenger (paise) Revenue (Rs) per trip EPK (paise) Estimated load factor Kote to Ramchandara Puram: limited stop (SETWlN) ordinary (APSRTC) The evidence from this particular route is that users are not sufficiently attracted by the higher quality service of the SETWlN buses to pay the additional fares. In August, 1980 the EPK of all SETWlN buses was 151 paise as compared to 281 paise for the APSRTC city service. Users of SETWlN buses do enjoy a higher quality service on the route surveyed. Comfort (measured by lower load factor) is higher than on APSRTC buses and speed is very much better. On this particular route, which includes large sections of rural and suburban roads vehicle trip times were about 20 per cent less by SETWlN than by APSRTC. Unlike the DTC limited stop services, which are only operated in the peak, the equivalent SETWlN services, operating throughout the day, should be able to make use of these savings in trip times through higher vehicle utilisation. Indeed, comparative figures for August 1980 show that SETWIN vehicles achieve almost 15 per cent more daily kilometrage than do APSRTC city service vehicles. The lower revenues generated per km or per trip, shown in Table 5, are to some extent off-set by a higher turnover or output. Whereas earnings per trip were approximately 80 per cent more for the ordinary service, in terms of earnings per hour of operation the difference is approximately 50 per cent. Even so, it is clear that revenues are not being maxirnised on this special service. 2.2 User characteristics lncomes: Table 6 shows the cumulative distribution of passengers' income for those using the high quality services in Delhi. Non-earners (mainly housewives and students) are excluded from this Table. TABLE 6 Cumulative income distribution of peak time passengers using high quality bus services, Delhi Monthly income (Rs) Limited stop Charter bus Ordinary Average (Rs)

10 The cumulative passenger income distributions of the limited stop and charter bus services are similar to each other and different from that of the ordinary bus users.. As might be expected the two high cost and high quality services seem to attract users with higher incomes; only about 40 to 50 per cent of therfi earn less than Rs 0 per month, whereas 70 per cent of users of the ordinary service are in this category. Table 7 shows the cumulative distribution of passengers' incomes for those using the high quality SETWlN buses in Hyderabad. TABLE 7 Cumulative income distribution of passengers using SETWlN limited stop bus services, Hyderabad Monthly income (Rs) Limited stop (SETWIN) Ordinary service (APSRTC) i Average (Rs) There is little difference in the cumulative passenger income distributions of users of the limited stop and ordinary services. It might have been expected that, as in Delhi, the differences in service price and quality would be reflected in a more marked difference in passenger incomes. The reasons for the similarity in distributions are perhaps a reflection of different journey purposes for which each type is being used. This is discussed further in the next section Journey purpose and trip frequency: As most passenger interviews were conducted during peak travel times, the overwhelming journey purpose was inevitably for employment purposes. In any case this would be expected for both the charter bus and limited stop services of Delhi, which are only operated in peak periods specifically for commuters. A split in type of service used between inward and outward journeys was observed on the limited stop services in Delhi. Regular users of the morning service to the city centre would often use an ordinary or charter bus for the return journey in the evening. However, users of both the charter bus and ordinary bus tended to be regular customers on a daily basis in both directions. In Hyderabad SETWlN bus users made use of the limited stop service infrequently which may help to explain the income findings noted earlier. Whereas it might have been expected that users of SETWlN buses would come from high income groups and use the vehicle on a frequent basis, they more likely come from lower income groups and use the vehicle on an infrequent basis (particularly when no other option is open to them) Reasons for choice of mode: Passengers were asked to give their main reason for their choice of a particular mode. Table 8 gives the percentage distribution of the responses given by users of the high quality special services in Delhi.

11 TABLE 8 Distribution of passengers' reasons for choice of quality service, Delhi (percentages) No interchange (direct) Fast Regular Convenient/comfortable Seats available Uncrowded Other special service unavailable Other reasons Limited stop Charter bus Total Most of the expected qualitative improvements (over the ordinary bus service) are nominated by users of each service, who after all, pay extra for these benefits. A large portion of charter bus users grouped their reasons under the general heading of convenience, which tends to embrace all the other headings. In Delhi, passengers were asked to estimate their wa!ting time for their chosen service. Users of the high quality special services estimated relatively short wait times: the majority of those using a charter bus wait less than 5 minutes; the majority of those using a limited stop service wait between 15 to 20 minutes. By comparison users of the ordinary service estimated an average waiting time of 25 minutes. These waiting times have not been verified, but in the latter case the average is undoubtedly an overestimate. On six routes monitored in Delhi 2 the average wait time, even in the peak, was never greater than 20 minutes (although, of course, individuals do have to wait longer than the average). Perhaps those who think they use the worst service (the ordinary bus) feel they must justify their expectations and perceptions by overestimating wait times and downgrading the service. Users of limited-stop buses in Delhi claimed that they saved approximately 19 minutes (in total) using this service instead of the ordinary service. This saving was due to less waiting time before boarding the vehicle, not having to make an interchange (with its consequent delays) and by there being a marginally faster in-vehicle journey speed, especially for passengers travelling to Kendriya Terminal. This average time saving of 19 minutes cost the passenger an average 20 paise extra in additional fare costs compared with the ordinary service. Charter bus users probably make gains in trip times over both limited stop and ordinary service passengers. 3.1 Operational characteristics 3. SPECIAL TYPES OF SERVICE Only one example of this category has been studied, the railway special operated by the DTC in Delhi. The railway special services were introduced so that rail passengers could be provided with direct routes from the railway stations to various residential areas of Delhi. The DTC operates five railway special services from the suburbs of Delhi to both Old and New Delhi Railway Stations. Four of these services are operated throughout the day from to whilst one (from Tilak Nagar to Old Delhi Railway Station) is operated as a peak time service only. The four services which are operated throughout the day link Nangloi, and Janakpuri to New Delhi Railway Station and lit (Hauz Khas) and Seemapuri to Old Delhi Railway Station. Passengers are allowed to transport 8 kg of luggage free of charge and can transport additional

12 luggage at modest cost. With this luggage facility the DTC hope to encourage potential taxi and auto-rickshaw passengers to use the service. A flat fare of Re 1 is charged for passengers (50 paise for children) so that the service is considerably cheaper than that provided by the auto-rickshaw or taxi, though the latter provide a door to door service which the railway special cannot match. A railway special bus is shown fn Plate 3. Table 9 shows average daily operational characteristics of DTC railway special services for December, TABLE 9 Daily o ~erational data of DTC railway special services Route No. Sch Buses Op Sch Trips Op Km Sch per vehicle Op Revenue per vehicle (Rs) EPK (paise) Source: DTC records, December 1979 All vehicles were operated on a daily basis throughout the month but a small number of trips (maximum of 5) were missed daily on each of the routes. Daily utilisation per vehicle ranged from approximately km which was higher than the average for the total fleet which was approximately 220 km. Vehicle and crew productivity was therefore appreciably higher than for DTC ordinary services. However, earnings per km (EPK) ranged from 95 paise to 152 paise; this compares with the average EPK for all DTC operations in December of 174 paise. Railway special services were not, therefore, as remunerative as the average for all routes operated and for route 6 the average daily EPK was only 54 per cent of the average for all routes. (The route was new, however, having only been introduced during December 1979.) As with the SETWlN buses (Table 5), the lower earnings per km and per trip are partly, but by no means totally, off-set by the additional vehicle utilisation. During February 1980 two railway special services were monitored over a four day period to obtain operational data such as service frequency, trip times, revenues and patronage levels. The services monitored were lit (Hauz Khas) to Old Delhi Railway Station and Janakpuri to New Delhi Railway Station, services 2 and 3 in Table 9. Monitoring was undertaken over several days and at different times of day to give a wide range of observations throughout the operating day. Table 10 gives results from the surveys, together with comparative data for the whole of the DTC city network for the year 1980/81. TABLE 10 Comparative earnings on railway special and ordinary services, Delhi Route Average bus load per trip Average fare (paise) Revenue per trip (Rs) Revenue per bus per day (Rs) EPK (paise) Estimated load factor Railway special No Railway special No Total DTC city service* * Source: DTC operating statistics for the year

13 The evidence from this Table is that despite lower earnings per km, it is still possible for buses on one railway special route to gross as much in a day as the average ordinary service buses. It is not entirely clear why railway specials should achieve such good utilisation figures, which help bolster their daily revenues, since their operational speeds are not outstanding by DTC standards. On routes 2 and 3 the average vehicle speeds were about 22 kph. 3.2 User characteristics lncomes: Table 11 shows the cumulative distribution of passengers' income for those using the railway special services and the ordinary services in Delhi. TABLE 11 Cumulative income distribution of passengers using railway special services, Delhi Monthly income (Rs) Railway special Ordinary Average (Rs) The rail special, though charging a higher fare than the ordinary service, would appear to be attracting users with a similar income. This might be expected if the rail special is fulfilling its main role, ie. providing a cheap alternative to the taxi for the infrequent journey to or from the station Journey purpose and trip frequency: An extensive survey (which included off-peak observations) was made of railway specials to see whether the service was being used for the purposes for which it had been designed, ie. to provide a special service for those carrying luggage to or from the railway station. On the two routes monitored only about 20 per cent of passengers were using the railway special in the above manner; over half the passengers were travelling to or from work. Furthermore, these passengers were undertaking journeys by rail special on a frequent basis, often daily, although not necessarily in both directions. (Timings of the railway special may not fit in with work timings, and therefore alternative bus services may have to be used for one of the daily work-purpose journeys.) A significant number of traders used the railway specials in Delhi to transport their wares, perhaps making use of the free carriage of luggage concession on this service Reasons for choice of mode: As with the high quality services passengers using railway specials were asked to give their main reason for their choice of this service. Table 12 gives the percentage distribution of their response. Users of the railway special did not refer to the specific purpose of the service; they seemed to value the service in very much the same way as did users of the limited stop and charter services. Perhaps this reflects the fact that the service tends to be used by commuters, rather than those travelling to and from the railvcay station. 10

14 TABLE 12 Distribution of passengers' reasons for choice of railway special, Delhi (percentages) No interchange Fast Convenient/comfortable Seats available Uncrowded Other i5 13 Most passengers of specialised services make comparisons with the ordinary bus service, since most have transferred from the ordinary to the special bus service on the introduction of the latter. For example, 82 per cent of users of the railway specials previously used the DTC ordinary bus service, while only 14 per cent would have used the auto-rickshaw or taxi, had the railway special service not been functional. Like users of the high quality services, users of the railway specials estimated relatively short wait times, with an average of 7 minutes. This compares with the average waiting time for ordinary bus users of 25 minutes, as already noted. 4. SERVICES TO ATTRACT CAR USERS TO PUBLIC TRANSPORT 4.1 Operational characteristics The luxury, air-conditioned bus service introduced by BMTA in Bangkok in 1976 is examined in this category. It could equally well have been described in Section 2 since the vehicles also provide a high quality service at high cost. However, there is the additional aim of trying to encourage some car users to use public transport. This would appear to be a relevant and worthwhile objective in Bangkok which, unlike Delhi and Hyderabad, suffers from extreme congestion caused by private motor vehicles. The service was initially operated by using vehicles hired from the former illegal inter city services which were operated in various parts of the country. Soon after, an additional 200 new vehicles were purchased so that in April 1981 there were a total of 542 vehicles operating on 15 routes (ie. approximately 36 vehicles per route operated). In May 1981 a further vehicles were added to the fleet. Hates 4 and 5 show an air conditioned bus and an ordinary service vehicle operated by the BMTA. The existing route network (1981) of the air-conditioned service is rather inefficient since the buses tend to be operated on routes which operate between the present routes of the ordinary bus system thus Filling the gaps left by the ordinary system. In the future however, vehicle routes will be along high demand corridors so that passengers will have a choice of service provision between the ordinary and air-conditioned buses. This should have the effect of increasing the level of patronage on the air conditioned services. The average route length is 25 kin, With a range of between 8 and 45 km. Fare levels for the air-conditioned services are more expensive than for the ordinary service. There is a flat fare for the ordinary service of Baht 1.50* which enables the passenger to travel up to a distance of 12 km; for each additional 12 km the extra cost incurred * In April 1981 the sterling was approximately equivalent to Baht 44 (Baht 1 = 2.2 pence). 11

15 is Baht 1. In comparison the flat fare for the air-conditioned service is Baht 5 but this enables the passenger to travel up to 8 km only; for every additional 4 km the fare is increased by Baht 2 up to a maximum of Baht 15. The service is very comfortable and is supposed to be seating only, though on many occasions during the peak travel period standing passengers were observed. Even where standing is allowed the service is obviously far superior to that of the ordinary bus. Table 13 shows comparative operational data for air-conditioned and ordinary bus services operated by the BMTA. These data are for the operational year 1979/80. TABLE 13 Daffy operational characteristics ( ) of bus services operated by the BMTA, Bangkok Average daffy Average revenue Average fare Daily km EPK load per bus per bus (Baht) (Baht) per bus (Baht) Air-conditioned Ordinary Source: BMTA Note: Prior to March 1980, the fiat fare was Baht 1 ; this was increased to Baht 1.50 on March 1st, Unlike the examples of specialised bus service already examined, there are significant differences in the costs of the two service types operated in Bangkok. Air-conditioned vehicles, being more sophisticated than ordinary buses, are more expensive to purchase and maintain; in addition there are extra operating costs due to the air-conditioning unit. The BMTA have estimated their losses on each vehicle type as 96 Baht per bus per day for the ordinary service and 671 Baht per bus per day for the air-conditioned service. Whether the routes are poorly planned or the fares too high it is evident that the air-conditioned buses are a poor proposition for BMTA, and with load factors of less than 0.4, it is also likely that these vehicles are not very successful in attracting car users to public transport. Indeed, whether they are attracting any car users at all is examined in the next section. 4.2 User characteristics In Bangkok, passenger surveys were only carried out on air-conditioned buses, and therefore comparisons with ordinary bus passengers are not possible. The majority x)f passengers interviewed using the air-conditioned service (76 per cent) earned between Baht 2000 to 8000 monthly with the largest group of passengers earning between Baht 3000 to Only 15 per cent earned less than Baht 2000 and 9 per cent earned in excess of Baht 8000, the average monthly income being Baht Approximately equal numbers of passengers used the service once or twice daffy (36 per cent) and infrequently (34 per cent). The remainder tended to use the air-conditioned service once every two to four days. In response to a question about regularity of use for the journey being undertaken, 59 per cent stated that they always used the service whilst 41 per cent stated that they did not. Of the latter, most said they travelled by ordinary service, though a significant number selected the first bus to arrive, be it ordinary or air-conditioned. Over 90 per cent of users of air-conditioned buses claimed that this was the most convenient mode. 12

16 While the air-conditioned buses were largely introduced to attract car users on to public transport, over 80 per cent of users of the air-conditioned buses previously used BMTA ordinary services. Some 11 per cent of users have been attracted from their private cars. Similar results were obtained in response to the question "How would you travel if air-conditioned buses were not operated for a few days?". Passengers previously using cars would return to this mode, rather than use an ordinary BMTA bus. Passengers waiting for the air-conditioned bus service were asked to estimate waiting times for both the BMTA air-conditioned and ordinary service. In order to check their estimates with actuality, waiting times were monitored, passengers being timed from the moment they arrived at the stand to when they boarded a vehicle. The average estimated wait times were 10 and 10.5 minutes respectively for the air-conditioned and ordinary services. The observed waiting times showed that passengers had generally overestimated time, for the airconditioned buses by 20 per cent and for the ordinary buses by 50 per cent. There would appear to be an inbuilt prejudice against the service which has been rejected, a prejudice which in this case was not justified. 5. SERVICES WHICH IMPROVE THE OPERATORS PRODUCTIVITY 5.1 Operational characteristics In both the first two categories described in sections 2 and 3 a by-product of the improved quality of service or provision of special purpose service was the possibility of improved vehicle utilisation (over and above that of the ordinary service). It is possible however, that the main purpose of a specialised service might be the improvement of vehicle output. Express services offer reduced trip times to users, but unlike the limited stop services described earlier, they do not provide added comfort (for example, through restrictions on standing) and they incur only minimal (if any) extra charges over the ordinary service. In Hyderabad, APSRTC provides a small number of express services on heavy traffic corridors of the city in addition to the ordinary services operated. Services are provided on seven routes by a fleet of 43 vehicles (August 1980) which is about 7 per cent of total fleet strength. Additional vehicles are to be added as and when the service is expanded to other areas of the city. Express bus fare levels are an additional 5 paise per stage distance over ordinary APSRTC services. No surveys were undertaken of the performance of these express buses and there is only a limited amount of information available from APSRTC records. In August 1980 express vehicles were achieving better output than ordinary buses by about 10 per cent. This increased productivity was off-set by a much lower earning potential. The EPK for express buses was 204 paise as against 281 paise for ordinary buses. During a working day the ordinary bus could bring in up to 20 per cent more revenue than the express bus. The reason for such a large difference is difficult to understand, since it is unlikely that the minimal fare differential would deter many persons from using the express service. Some officers within APSRTC felt that the service was in its initial phase of development and still not reliable enough to make any positive judgement as to its success. 5.2 User characteristics Table 14 shows the cumulative distribution of passengers' income for those using the special services in Hyderabad. Again, non-income earning passengers are excluded from this Table. 13

17 TABLE 14 Cumulative income distribution of passengers using special bus services, Hyderabad Monthly income (Rs) Express (APSRTC) Ordinary Average (Rs) The cumulative distributions of passenger incomes for ordinary and express services are almost identical. This supports the notion that the marginal difference in fare levels has little influence on the choice of either service and that users of both services would tend to come from similar socio-economic backgrounds. Users of express buses were interviewed on their reasons for use of this service. Not surprisingly almost 80 per cent said they preferred the express service to the ordinary service because it is faster. The ordinary and express services were used twice daily by most passengers interviewed. 6. DISCUSSION Though the specialised services described have been classified as being typical of one of four categories it must be evident from the survey findings that these categories are not clear-cut. For example, most of the services described achieve some improvement in vehicle productivity, be it the prime objective (as in the case of the fourth category - express buses) or not, and while an express bus may meet its prime objective of improved vehicle utilisation, this must obviously be weighed against other objectives like operator's profitability and user satisfaction. In this section the special services are discussed in respect of these overriding objectives ie. operator's profitability and user satisfaction. Clearly it would seem illogical from the operator's point of view to introduce a special service if his profitability declines (or more profit could be made by employing the resources elsewhere). It might also seem illogical from the community point of view if resources allocated to a special bus service, for those who can afford it, detracted from the service offered to the vast majority of travellers who can ill afford high fares. A full cost-benefit analysis would be required to examine the benefit trade-off between different sections of the community and the costs of providing different options. For the purposes of this discussion the analysis is much simplified. Using the case-study material presented, some of the elements of a profitable service (which must be foremost in the operator's mind when introducing a service) are examined. Secondly, whether the network patronage as a whole has been affected by the service introduction (assuming a switch in resources from one type of service to another) is assessed. As a crude proxy it is contended that the service is of little value if it is less profitable to the operator than his ordinary service and if the network patronage has also been reduced (or not expanded optimally). Under any other conditions (extra profitability and/or extra patronage) then the service clearly has some merits which are probably worth pursuing. This analysis is simplistic; it ignores the benefits obviously gained (and attested to in the case studies) by those using the special services and, in the case of Bangkok, the benefits of reduced car congestion. However, it avoids the complex issues of cost benefit analysis 0ike valuing time savings) which have still to be resolved. 14

18 Apart from the air-conditioned buses in Bangkok, it is unlikely that the unit costs of operating buses on either an ordinary or a special service are vastly different. (Speed differences will have effects on fuel consumption, rate of wear etc, but these effects are not considered in this discussion.) Total daily operating costs will be different between the two services mainly because of any differences in vehicle utilisation. For example, using a simple cost model developed for DTC 2, it can easily be shown that a 20 per cent improvement in vehicle utilisation over the norm for DTC ordinary service buses of about 220 kin, gives rise to an increase in daily costs (direct and variable overheads) of about 10 per cent. This cost must be recouped (if the service is to be equally or more profitable than the ordinary service) by additional revenues generated through a higher turnover of passengers and/or a higher fare level (which if significantly higher than the ordinary fare will deter some patronage). It might be expected that the fare could be held constant (or marginally increased) and that, if load factors hold for the additional output, then revenue would rise by the same proportion as the increase in vehicle utilisation. This has clearly not happened in the case of express buses in Hyderabad. There, a 10 per cent improve- ment in vehicle utilisation together with a small increase in fare has been accompanied by a 20 per cent reduction in revenue. More generally the total revenue earned per day per bus (R) can be represented by the following identity: R = Load factor x Av. fare x seats per bus x vehicle km Passenger lead... (1) where the load factor is the ratio of passenger km to seat km and passenger lead is the average passenger journey distance in km. In Hyderabad the increase in vehicle km and average fare have been off-set by a decline in load factor. Most likely this is a reaction to the service itself and not the fares increase which is marginal; the improvement in vehicle speeds through the reduction in numbers of stops has not sufficiently compensated those who have to walk longer distances to the remaining bus stands. On the high quality and special services, where substantially higher fares were charged, the improvements in vehicle utilisation were of the order 15 to 20 per cent for limited stop services (in both Hyderabad and Delhi) and as high as 35 per cent on some of the railway specials in Delhi. St~ch large increases in vehicle utilisation, together with up to 150 per cent increases in fares, were off-set by a reduction in available capacity (standing being banned) and a decline in load factor partly due to the capacity restriction and partly to the fare increase. The load factor could be as much as 40 per cent down on ordinary services, while available vehicle capacity might be per cent down. With so many components of the revenue identity working in opposite directions it is not easy to forecast the outcome. Some of the rail special and limited stop services in Delhi did achieve higher earnings than an ordinary service vehicle, with sufficient to cover extra costs, while others did not. The limited stop service in Hyderabad clearly failed. In the case of the air-conditioned buses in Bangkok their unit costs of operation were significantly different from those of the ordinary buses. Vehicle utilisation was poor, fares much higher, and, no doubt as a result of the high fares, patronage was substantially lower than on the ordinary buses. The air-conditioned bus service makes much greater losses than does the ordinary bus service. In the case of all these services where fares are considerably greater than on the ordinary service, the responsiveness of patronage to the size of fare differential is clearly critical to the success of the operation. The fare differential is also clearly related to the locality of operation: it is obvious that high cost and quality services should not be operated to low income areas. One reason why some rail specials are not being operated 15

19 optionally is perhaps because the market for their design purpose does not exist in any quantity; the vehicles are picking up some commuter traffic, but unless this is substantial (which might result from the route being located through high-income corridors) then the vehicle cannot sustain high revenues. The evidence on patronage levels is limited but tends to suggest that in all cases there was a reduction on what might have been expected from an ordinary bus service. The limited stop services in Delhi handled only 50 per cent (at most) of the patronage (passengers per hours operation) of other ordinary peak time services. Throughout the entire operating day a DTC vehicle on ordinary service might handle around 90 to passengers per hour. On the railway special routes monitored passenger handling averaged about 30 to 40 per hour. In Hyderabad on the route monitored the limited stop vehicle had a passenger throughput almost one half that of the parallel ordinary service. Using the criteria discussed earlier it would appear that few of the special services described have much value to their operators or to the wider community of bus users, though individual users of the specialised services cleady do benefit. However, some of the specialised services were financially worthwhile and no doubt if the service were properly planned (getting the right balance between service level, fare and locality served) there is no reason why this should not be so in a greater number of cases. However, there is evidently a need for operators to know more about their markets and the responsiveness of patronage to changes in services on offer. The market is not homogeneous and suitable specialisation in service provision could bring rewards. However, in the wider context of community transport needs when supply is insufficient to meet demand the wisdom of maximising route profitability at the expense of route capacity may be questioned. There could be a compromise position with, on the one hand the provision of high cost and quality services to those who can afford high fares and, on the other hand, more high output express services at no extra cost to the user. 7. SUMMARY AND CONCLUSIONS Many stage bus operators in cities of the Third World are developing alternative or special bus services. These services are operated to meet perhaps one of four basic objectives: to maximise revenue by providing for those passengers who are willing to pay premium prices regularly for a high quality service; to provide a distinct and special type of service which may incur a high cost of service provision but which competes with taxi-like services, providing a cheaper alternative; to increase the productivity of vehicles and crews, through for example, reducing the number of request stops; to attract potential passengers who currently travel by personal vehicles and so to reduce traffic levels within the city. Special services generally offer the consumer some improvements on the normal stage - carriage service, for which extra payment is charged. In the case of productivity improvements (through, for example, the provision of express services) not every traveller may benefit; users will spend less time in the vehicle, though some may have to walk further to a bus stand than they would have done if using a normal stage carriage bus. The rationale for charging marginally higher fares on these services is not totally clear. If productivity improvements outweigh any loss in custom (and hence revenues), then there should be no need for fare differentials between the two services. However, it may be expedient for identification and marketing purposes to charge different rates for the different services. II Examples of all four types of special service were monitored in one or other of the three cities selected for observation in order to see how well they met these various objectives. Two special services operated by the DTC in Delhi were included: the limited stop service, a quality service at premium fares and the railway specials catering for non-regular users with luggage again at premium fares. In Hyderabad the APSRTC were operating express services which were marginally more expensive than the ordinary service. In the same city SETWIN were operating limited stop services of higher quality (than both the express and normal service of APSRTC) at premium 16

20 rates. In Bangkok the BMTA were providing a very high quality service, employing air-conditioned vehicles, at premium fares. Quality differences in services between the special and the normal service were not always so obvious as in Bangkok. For example, the buses used by DTC on its limited stop service were no different from those used on ordinary services. The difference in quality arose from the restrictions placed on numbers of passengers carried. (Every passenger was supposed to be guaranteed a seat.) In the case of the limited stop services operated by the DTC there was some evidence that revenues on some routes could be substantially above those earned by ordinary bus services. Account was taken of the fact that limited stop services were only operated during peak periods when earnings might be expected to be high. Vehicle speeds were much improved over ordinary services. However, because routes to some parts of the central area tended to be more direct by ordinary service, in-vehicle times on limited stop services were not always competitive. It was probably no coincidence that the highest loadings on the limited stop service were achieved on route sections where journey times were better than by normal service. Many users preferred the limited stop services because it gave them a shorter journey time overall. The quality service operated by SETWlN in Hyderabad was performing badly in comparison with the normal stage-carriage service at APSRTC. The service was operated throughout the day, and the routes were long. It may be that the market for this type of service exists only during the peak when the availability of seats is at a premium and users are prepared to pay a high price for this facility. Evidently the charter bus operators in Delhi work on this principle, since they only provide commuter services during the peak periods, employing their vehicles on other duties in the off-peak. In this respect a problem for the regular stage-carriage operator is that during the peak there will be great pressure to carry as many passengers as possible with the limited stock of vehicles. To divert buses to the quality market is not likely to be well received by the majority of commuters if this means any reduction in their service. The express services operated in Hyderabad by APSRTC in parallel to their own normal stage-carriage operations seemed to be meeting the objective of higher vehicle productivity. However, the additional 10 per cent output from the vehicles was lost in lower revenues, compared with the normal service, and this despite the marginal fare differential in favour of the express service. The reason for this difference in response could not be identified from the surveys. Users of both services came from a similar socio-economic background and made use of this preferred mode on a regular basis. In-vehicle journey times were less on the express services, and this was the reason usually given by users for selecting this mode. It could be argued that most travellers would not be too concerned about marginal in-vehicle time savings and would be more likely to pick the first available vehicle (be it express or ordinary) in order to minimise waiting time. However, the fact that users of each service tended to be regular users suggests that they preferred to wait for their chosen mode. The differentials in loading may be a reflection of the fact that the express buses were not being operated from high-demand bus stands, though it seems unlikely that the APSRTC would have overlooked this possibility. Cost was unlikely to play a major part in choice, since the fare differential was small. Evidently some combination of all these factors was at play and perhaps more careful attention to the choice of route in which to run express services is required. The experience of APSRTC in Hyderabad in the operation of express buses is not encouraging, though it may not be typical. The luxury service being operated in Bangkok by BMTA has proved to be a high loss-maker in comparison with the normal service. The output (daily km) of the average luxury bus was some 6 per cent less than the average for an ordinary bus, and 80 per cent fewer passengers were carried per bus. Despite this the earnings of the luxury bus were higher than those of an ordinary bus, though not sufficiently high to cover the additional costs of running luxury vehicles. The latter were 75 per cent more expensive to run (per km) than the ordinary 17

21 buses. As with the SETWIN buses in Hyderabad, these luxury buses were being operated throughout the day; surveys showed little demand for the service in the off-peak period. If the aim of the BMTA in operating luxury bus services was to encourage car-owners to use public transport (instead of their own vehicle), then they have not been particularly successful. Over 80 per cent of users interviewed were attracted from existing ordinary bus services operated by BMTA. Only just over 10 per cent formerly used a car. There was some suggestion that the route network for BMTA luxury buses was at fault and that some rationalisation in the future will improve the financial performance of the vehicles. There may also be a good case for both BMTA and SETWIN (in Hyderabad) to introduce fare differentials on their luxury buses, between peak and off-peak, in order to improve loadings in the latter period. The remaining example of a specialised bus service monitored in these surveys was the railway special operated by the DTC in Delhi. This was providing a specialist service to persons needing conveyance to and from the main railway terminals in the city centre. In practice it was evident that this service was being used by ordinary commuters on a regular basis. Only 20 per cent of passengers interviewed were using the vehicle for its intended purpose. Despite the circuitous routes taken by these vehicles, commuters were prepared to pay an additional fare (over and above that of the ordinary service) in order to get a seat. Over one in four users specified this (the ability to get a seat) as a main reason for using this service. As with the SETWlN and BMTA luxury buses, the railway specials of DTC were operated throughout the day. Their output in terms of passengers carried and revenues earned was not outstanding by comparison to ordinary services. Apart from in Hyderabad, all the quality services have attracted users from above-average income groups who can clearly afford the extra costs involved. For the most part, users of these special services have diverted from existing normal bus services. Only in the case of railway specials in Delhi was there any obvious diversion from taxis and auto-rickshaws (almost 15 per cent of users), though it is likely that these were the non-regular users of the service, ie. those for which the service was designed. Comfort and seat availability were two of the most important reasons given by users for choosing special services. Time savings were given little weight, except by users of the express services in Hyderabad. It is evident that most of the special services were liked by their regular users. However, few of the services were as profitable as comparative ordinary services, despite generally higher vehicle utilisation and higher fares. Furthermore, load factors on all the specialised vehicles were low; in an environment where there is not enough supply to meet the demand, it could well be argued that these vehicles were being wasted. The following conclusions are drawn: High quality services at premium rates were in demand during peak periods. Operators who tried to maintain the same service at the same fare levels throughout the day achieved poor financial returns. There may be a case for either restricting operations to the peak period (and using the vehicles on other duties in the offpeak) or charging differential fares between peak and off-peak. high quality services at premium rates were attracting above-average-income travellers who could clearly afford to pay higher fares on a regular basis. Most users were paying forcomfort and a seat, rather than time savings. most users of special services were attracted from existing ordinary services. In this respect the BMTA objective of trying to reduce the use of personal vehicles in Bangkok has not been a great success. The railway specials were not much more successful in attracting potential taxi and auto-rickshaw users. the market for railway specials was not very apparent. These services were being used mostly by commuters. 18

22 - express bus services, which do not involve any other quality improvements, were having little success in Hyderabad. Vehicle productivity was improved but each vehicle carried fewer users than an ordinary service vehicle. It was not clear why this was happening. - judged on the criteria of either profitability or passenger handling, few of the services examined had much merit in comparison with ordinary bus services. However, users clearly found the special services satisfactory. More careful planning could probably improve the performance of many of the services. To sum up, a market probably exists for a high quality service at premium rates, during peak periods. Undoubtedly this is an expression of the general lack of supply within the public transport system during that period. The problem for the hard-pressed stage-carriage operator is that he may find it difficult to divert large resources to this specific market since the majority of travellers will still be dependent on the cheap ordinary service. It is of interest to note that over the period I the DTC have adopted a policy of introducing a limited stop bus service on each route operated. 8. ACKNOWLEDGEMENT The work described in this report forms part of the joint research programme undertaken by the Indian Association of State Road Transport Undertakings and the Overseas Unit (Unit Head: Mr J.S. YerreU) of the Transport and Road Research Laboratory. 9. REFERENCES. FOURACRE, P R, D A C MAUNDER, M G PATHAK and C H RAO. Public transport supply in Indian cities. Department of the Environment Department of Transport. TRRL Report LR 1018, Crowthorne, 1981 (Transport and Road Research Laboratory).. FOURACRE, P R, D A C MAUNDER, M G PATHAK and C H RAO. Studies of bus performance in Delhi, India. Department of the Environment Department of Transport, TRRL Report SR 710, Crowthorne, 1981 (Transport and Road Research Laboratory). 19

23 _ Plate 1 A "Limited stop' bus operated in Delhi by the DTC Neg. no. B180/81 J'... 3: I Plate 2 A SETWIN single deck vehicle operated in Hyderabad Neg.no. B557/81

24 Plate 3 A 'Railway Special" bus operated in Delhi by the DTC

25 Neg. no. R315/81 Plate 4 An air conditioned bus operated in Bangkok by the BMTA Neg. no. R315/81 Plate 5 An ordinary service bus operated by the BMTA in Bangkok (2858) Dd ,400 12/83 HP Ltd So'ton G1915 PRINTED IN ENGLAND