India - Report on Bombay (Vol. 4 of 5) : Annex III : transportation

THE WORLD BANK URR7306 DEVELOPMENT ECONOllICS DEPARTMENT VOL. 4 URBAN AND REGIONAL ECONOMICS DIVISION URBAN AND REGIONAL REPORT No. 73-6 REPORT ON BOMBAY ANNEX III: TRANSrORTATION B. B. KING and Others l!arch 31, 1971 These materials are for internal use only and are circulated to stimulate discussion and critical comment. Views are those of the author and should not be interpreted as reflecting the views of the World Bank. References in publications to Reports should be cleared with the author to protect the tentative character of these papers.  REPORT ON BOiDBAY ANN_EX III - TRANSPORTATION TABLE OF CONTENTS Page No. I. COMNNUTER PASSErNCER RAIL SERVICS ............1 Introluction ......... 1 Existing Facilities, Services and Traffic ... 1 Fares. Earnings and Costs ..... .............. 5 Studies of Future Requirements .... .......... 7 Current Improvement Programs and Proposals 10 Observations on Mass Transit Studies aad Proposals ....... .......................... 14 Alternatives to Consider ..... ............... 16 Iprovements of Existing Services .... ....... 18 II. RAIL GOODS TRAFFIC ..... ..................... 21 III. BUS TRANSPORTATION IN BOMBAY .... ............ 24 The Operating Company ..... .................. 24 Route Structure and Schedules .... ........... 24 The Fleet ....... ............................ 25 Road Traffic ....... ......................... 26 Capacity and Growth ..... .................... 27 Fare Structure ...... ........................ 27 Traffic and RevenIuCs ..... ................... 29 Capital Requirer!ents and Financing .......... 29 B;onus Incentives ...... ...................... 30 Types of Buses ...... ........................ 31 BES&T and the Governmenlt ..... ............... 32 Demand or Supply? ...... ..................... 33 Cross Subsidies and Misallocations .......... 33 Mtanagement ....... ........................... 34 Prospects for the Future ..... ............... 35 IV. BOWBAY ROAD DEVELOPMENT PLAN .... ............ 39 Work in Progress ...... ...................... 40 Road Program Issues ..... .................... 40 V. AIR lRANSPORI'Al'ION ...... .................... 45 Introduction andl Background .... ............. 45 Domestic Passenger Travel ................... 45 International Passcnger Traffic ... ......... 45 International Freight and Mail 'l'raffic ...... 46  TABLE OF CONTENTS (Cont 'd) Page No. Domestic Freight and Mail Traffic .... ....... 46 Aircraft Movements ........................ 46 Airport Current Finances .................... 47 Bombay Airport Modernization .......... I ..... 48 VI. PORT OF BOMUAY .............................. 52 Introduction ... ....... .......... 52 Administratlon ........ ...................... 52 Finance ....................-. 52 Port Facilities ........ ..................... 52 Port Traffic ......... ....................... 55 Port Trust Railway .................., 56 Current Developments ...... .................. 56 Port Haster Plan Study ...... ................ 57 New Port Development .........., ........... 57 Transport Modifications at Existing Docks .., 59 LIST OF TABLES 1. Growth of Suburban Passenger Traffic 2. Transit P4ssenger Trips Crossing Screenlines (Bus and Rail) 3. D4ily Transit Passenger Assignments (1981) 4. Daily Average Passengers Carried by BES&T, 1966/67 to 1969/70 5. Construction Recommended by Bombay Traffic and Transportation Study 6. Bombay Road Program 7. Bombay Road Works in Progress 8. Bombay Road Program - Recommended Stage Development  REPORT ON BOMBAY ANNFX III - TRANSPORT'ATIoN I. CODIUTER PASSENGER RAIL SERVICES Incroduction 1. Bombay's commuter railroads are simultaneously benefactors, vic- tims, and villains. The city cannot function today without them; there is no other way to carry the daily surges of workers from home to office and factory and back again. For their efforts to carry these ever-increasing masses, the railroads draw more criticism than praise: complaints about overcrowding In the trains, deteriorating quality and reliability of serv- ices, shocking accident records, and the like. Their patrons, on several occasions, have expressed their dissatisfaction by burning the trains. While some of the complaints may be unjust and the ways of voicing them extreme, there are certainly symptoms of malaise in the railroads. unfortunately, the railroads, through their service and pricing policies - with a helping hand from regional geography and a distorted urban land market - have helped to foster the pattern of urban development which is one of the sources of their troubles. Unrealistically low rail fares and the availability of rela- tively cheap and uncontrolled suburban land have encouraged the location of residences farther and farther from places of work, thereby producing ever- longer work trips and ever-greater demands on the rail system. In addition to creating many of the problems which now beset the railroads, this suburban growth has allowed decision makers to postpone a confrontation with the re- development needs and opportunities within the city. Particularly now that proposals are being put fo:r for major new rail transit facilities, it is important that they be viewed in the context of overall urban development needs, plans, and resources. Existing Facilities, Services and Traffic 2. There is no single and separate entity responsible for providing rail commuter services in lombay. Both the Western Railway and the Central Railway, which have their headquarters In Bomhay, have developed commuter services as an adjunct to their main line passenger andl freight services. rhe Western Railway main line comes from Ahmadabad ar.d enters the Greater Bombay area at the north, crosses Bassein Creek, runs the full length of the island near its western edge, and terminates at Churchgate Station in the Fort Area. The suburban tracks of lWestern Railway parallel the main line and extend 34 km from Ghurchgate Station to Borivli; from there, com- muter trains use the main line tracks to Virar, another 26 km to the north. The Central lRailway, coming from the two crossings of the Ghats near Nasik and Poona, approaches Bombay from the northeast through Kalyan and rhana. Separate commuter tracks start at Kalyan and parallel the main line tracks to Kurla where the commutcr lines split into two branches which both event- ually end at Victoria Terminus. The main line and one commuter branch gener- ally follow the main axis of the island, and the 15 km llarbour Line branch skirts the eastern shoreline before reaching Victoria Terminus. There is  an additional 6 km feeder line from the Trombay area to Kurla, and another 2 km connection from Raoll Junction, on the Harbour Line, to Mahim Station on the Western Railway line. This is the only track link between the Western and Central Railways, although passengers may cr-nect from one railway to the other through a common station at Dadar. Kalyan to Victoria Terminus is 54 km; beyond Kalyan, the Central 'Railway operates commuter trains on the two main line tracks to Kasara (67 km further) and KarJat (46 km further). In total, the "commuter-shed" for Bombay extends 60 km to the north, 121 km to the northeast, and 100 km to the east and southeast. All the suburban com- muter operations for both the Central and Western railways are on electrified broad-gauge track. Both railways, however, operate meter-gauge and narrow- gauge in their nonsuburban systems. 3. The distinction between main lines and commuter lines is more a matter of nomenclature than practice. From Borivli to Virar on the l1estern Railway and beyond Kalyan on the Central Railway, all trains run on the same lines. Between Borivli and Churchgate Station or Kalyan and Victoria Term- inus, inter-regional trains do use only the main lines but commuter trains run on both the main line and commuter tracks in order to provide greater service capacity. An average of about 15 long distance main line trains operate each direction daily at Bombay Central, the main line terminal for Western Railways, and 21 at Victoria Terminus. The present procedure is to permit "fast" commuter trains - i.e., those which do not stop at all suburban stations - to use main line tracks jointly with scheduled main line trains during peak traffic periods. The Western Railway run!: approximately 469 commuter trains per average weekday. l/ This is a fairly high frequency, considering that the Churchgate tu Grant Road section is only double tracked and a turn-around is required at Churchgate. The Central Railroad operates 352 trains on the major line and 238 on the Harbour Branch, plus 100 trains in the Kurla and Kalyan areas. 1/ The high frequency of suburban commuter ser- vice on both railroads makes it difficult to operate main line freight and passenger services in Bombay on tracks which are used in common. The rail- roads would like to segregate t.e quadruple track, using a double track for suburban service and the other double track for mainline services. In view of the substantial congestion on the railroads at the present time, and probably the achievement of capacity in terms of headways -- at least at the peak hours, it does not seem appropriate to segregate the tracks in this manner. The effect that the present mixed use of tracks has on passenger capacity has not been precisely determined but the presence of main line trains (and the priority which they apparently have) is certain to reduce the capability for carrying comiuters below the potential capacity of a purely commuter service - a point which is discussed further in relation to possibilities for future improvements. (para. 43.) 4. Table 1 shows the growth in daily commuter passenger traffic on the Western and Central Railways from 1950 to 1967, and the corresponding changes in numbers of trains run. The total daily traffic now exceeds 2.2 1/ 1967 data.  -3- million passengers and is rearly equally divided between the two railways. An even more impressive figure is that, of the 2.2 billion total passengers carried by all Indian Railways in 1967 (the (late of the Bombay Traffic Cell studies), 35% were Bombay commuter service passengers. The growth of traf- fic for the 1Western lRailway, for which data were provided back to 1965-66, suggests atn annual growth rate for the period 1965-66 to 1969-70 of about 4 percent per year. While this is not very rapid growth, it should be noted that the growth in the period 1965-66 to 1966-67 was about 12 percent, and the next three years' growth in total is only about 5 percent. Similar pat- terns are seen to exist when individual, first-class, and third-class serv- ices are considered. While first-class services on the Western and Central Railroads runs about 10 percent of the total traffic, the growth rate of first- class service on the Western Railroad for the period 1965-1966 to 1969-70 has been 9 percent per year. In the first year of that period, it exceeds 20 per- cent. A similar pattern of growth appears on the Central Railroad. llowever, both roads appear to have declining traffic in 1968-69. The figures strongly suggest that the shortage of capacity is imposing a substantial constraint on traffic growth. 5. lTe railways have about reached the limit in numbers of trains they can operate during a peak hour, given their present facilities, equipment, and operating procedures. The growth is now being absorbed by increases in the average number of p.-zqengers per train and by spreading traffic over longer morning and evening peak periods. These periods now extend from about 8:30 to 11:00 a.n. and 4:30 and 7:30 p.m. Loads reach their absolute peaks at terminals in the Fort Area between 9 - 10 a.m. and 5 - 6 p.m., when some 90,000 passengers enter (a.m.) and leave (p.m.) the Fort Area each hour through four downtown terminals (Churchgate, MSarine Lines, Victoria Terminus, Masjid). 6. Peak period operations, as in all urban transportation systems, are the crucial considerations: they are the basis for establishing schedules and equipment requirements, they strain the capacity of lines and stations, and they largely determine the financial situation for ,he operators. Pre- sently the Western Railway runs 18 trains and the Central Railwav runs 26 trains durtng the peak hour. (The Central Railway figure includes both the Ceniter Island and llarbour Line branches.) Trains normally cons!st of 9 car rakes, although some old 7 and 8 car rakes are -'il! being used until they can be replaced by new stock. 7. The "Marked Carrying Capacity" of these trains depends on the num- ber of Class I and III l/ cars which are included and the particular stock being used; it usually ranges between 1674 and 1778 passengers per 9-car rake. Tlhis "capacity" is described as that which enables passengers to ride in tolerable standing comfort (plus, of course, the seated passengers). For 1/ The Central 11ailroad uses Class III and composite Class I-III coaches, the latter heing divided roughly two-thirds for Class I and one-third for Class III and having slightly lower capacity.  - 4 - the newest rakes, this marked capacity is 867 seated and 867 stariing pas- sengers, or 1,734 in all. Actual peak period loading practice bears little resemblance to this "marked capacity," however. The railways estimate that peak periocl trains carry 70% overloads, or about 3,000 passengers instead of 1,700. The 3,000 passenger figure would be more in line with peak period space utilization elsewhere; i.e., it is equivalent to full seat occupancy plus 1.3 sq. ft. per standing passenger. 8. Scarcely any information is available regarding the actual numbers of riders on peak period trains, aside from the estimated hourly arrivals of passengers and the known nunbers of trains at terminals in the Fort Area. For the W4estern Railways, the highest number of arrivals in tie Fort Area (at Marine Line Station) in a single hour of an average weekday is 45,000, or about 2,500 passengers per train. Of these, about 2,000 pas3engers per train debark at Churchgate Station. The numbers for the Central Railways are not strictly comparable since some rakes on the Harbour Line have 8 in- stead of 9 cars. However, 47,700 total passengers arrive at Masjid and Victoria Terminus during a single hour, about the same as on the Western Railway, and this load is spread over 26 rather than 18 trains. Assuming the same loading per car on both Central Railway Lines (which may not be the case), the average train load for an 8-car rake is about 1,700 and for a 9-car rake, about 1,900 passengers. These figures suggest that, although the marked caprrity is regularly exceeded, comparatively few trains on either railway arrive in the Fort Area with 3,000-passenger loads even during the highest traffic hour. 9. This is not to say that the railways are not having capacity prob- lems, but rather that all the problems are not clearly identifiable from existing information about passenger movements. Part of the difficulty stems from the fact that traffic studies have been largely concerned with the "Fort Area problem" and have tended to ignore conditions and travel pat- terns at other locations. There is evidence from the studies that, at least on a daily basli, the maximum rail-passenger traffic (peak load point) oc- curs not at the Fort Area but considerably further north on the Island, near Dadar. It is a reasonable expectation that traffic would start diminishing before reaching the last stations of a dead-end line. It seems particularly likely to be the case in the specific situation of Bombay where textile mills and other large industrial employers, commercial sub-centers and other in- tensive activities are located between the Fort Area and the major residen- tial areas in the northern suburbs. Virtually no data are available, how- ever, to describe passenger movements into, out of, and through these inter- vening areas during peak periods. 10. What is not clear from discussions of corridor capacities and train loading characteristics is that there are also problems related to passenger collectinn and dispersion at some stations in the Fort Area. Although Church- gate and Victoria Terminus are packed with people during the rush hours, pas- sengers are able to move quite rapidly through the stations themselves. The main problems, at the moment, are outside. People spill out of the stations and block the sidewalk with queues waiting for buses and taxis and disrupt movements of traffic on the abutting main streets.  - 5 - 11. Safety is another serious problem for the railroads. In the past year, more than 600 fatal railway accidents were re9orted in the Bombay area, mostly .ivolving either passengers falling from moving trains or pedestrians who were struck while walking on the traclks. Passenger accidents are larvely attributable to crowdeci cond(itionis on trains, the absence of doors on the ccaches, and the common practice of riding outsi4e the coaches on Lhe foot- boards (or even on top of coaches), frequently when the train is not even filled. The pedestrian accidents continue to occur despite the railways' efforts to keep persons off the tracks by building walls and fences at the rinht-of-way limits. If, in addition to the fences, pedestrian bridges were provided at more frequent intervals in denselydeveloped areas, this trespass- ing problem might be alleviated. However, many accidents occur at stations when passengers alight from trains (often still moving) and short-cut across adjacent tracks to board another train or to beat the crowd to the exits. Some reduction in these accidents might be achieved by barriers between lines, but a solution to the problem is probably more a matter of training and dis- cipline than physical controls. 12. Although not identified by the railroads in terms of being a prob- lem, certainly one of the important needs whlich should be considered is the replacement and modernization of rolling equipment. ?uch of the equipment used is very old. Of 383 electric multiple units, 116 are considered by the railroad to be over age (over 40 years old). The Central Railroad does not give specific ages for their equipment but 232 of their 448 units are clas- sifie(d as "new." The remainder are apparently quite old. The annual reports refer frequently to the maintenance difficulties with t'e 1928 stock and the difficulties of getting replacement parts because of foreign exchange limita- tions. Fares, Earnings and Costs 13. Six basic types of tickets are sold to commuters on the Bombay suburban railway services. The principal classifications are first class and third class. Within each of these classes, one can buy a single-journey ticket, a monthly season ticket, or a quarterly season ticket. For purposes of calculating passengers, Llte railroads assume that a monthly season ticket is equivalent to 50 Dassengers and a quarterly season ticket to 150 passengers. In terms of price, however, very substantial discounts are available for sea- son ticket holders. Monthly ticket holders may pay as much as 13 times the price of a sin,;le-journey ticket but probably use the ticket about 50 times. For very lobug trips, the monthly ticket price is only about 6 times the single-Journey fare. Additional liscounts occur when quarterly tickets are purchased1. The quarterly ticket is approximately 2-1/2 times the price of a monthly season ticket. For third-class tickets, the multiples are also fairly low, with monthly season tickets ranging from about 20 to about 8 times the price of a slngle journey. Again the quarterly season ticket Is only 2-1/2 times the price of the monthly season ticket. Not surprisingly, as a result of this type of fare structuire, about 2-1/2 times as many season tickets are sold as sin!le-Journey t!ckets for third class and about 20 times as many first-class season tickets are sold as first class single-journey tickets on the Central Railroad.  - 6 - 14. Several recent fare changes have taken place, but they were gener- ally o)f a minior nature. In 1968, third class monthly season tickets were in- creased in price by 25 palse, and first class tickets by Rs. 1. On the first of October 1969, quarterly season tickets were increased in price by 75 paise for third class tick!ts and Rs. 3 for first class tickets. The management of the two railroads indicate that fare increases are politically very difficult in Bombay and that it would be difficult to raise the fares sufficiently to offset the present estimated deficit from suburban services. The present fare structure has a very substantial taper. For monthly third class season tickets, the rate per kilometer tapers from Rs. 1.23 down to 20 paise per kilometer. For trips from the ;ost northern point in Bombay City served by the Central Railroad (Sion) to Victoria Terminus, the monthly season ticket is Rs. 6.05 and the price per kilometer is about 45 paise. For the trip from Kalyan Junction to Victoria Terminal, the third class monthly season ticket is Rs. 13.60 for a 54-kilometer trip, or about 25 paise per kilometer. In addition to the substantial fare taper and very pronounced discounts for season tickets, the railroad also provides student fares. First class, student season tickets are sold at half the monthly adult price. One might speculate as to why students should be permitted half fare for first class service, when they can travel third class at less than even half the first class fare. If the first class service is indeed a luxury, preferred type of service, there appears to be little reason to provide students with half fare for that quality. 15. The earnings from suburban rail service have grown somewhat faster than the number of passengers principa1ly as a result of the fare increases that have taken place in the last iew years. First class earnings on the Western Railroad, for example, have grown 14 percent annually from 1965-66 to 1969-70. Third class earnings during thle sae period have grown 12 per- cent annually, with total revenues in lakhs going from 554 in 1965-66 to 876 in 1969-70. Total re.-enues for the Central Railroad in 1968-69 from suburoan services were Rs. 772 lakhs, with about 1/8 of the revenue coming from first class passengers. 16. Prior to 1966-67, the railroads reported estimated allocations of cost and earnings from electric, multiple unit suburban *;ains. In 1966-67, the requirement to report such data was elirminated. Using 1965-66 as a guide, they earned 1.15 paise per passenger kilometer at a cost per passenger kilo- meter of 1.02 paise, while 10.6 rupees were earned per train kilometer and the cost per train kilometer was 9.51 rupees. Thus, in 19)65-66, according to the railroad allocations, the suburban train s2rvice generated a profit. No notes are available, however, in the reports, telling V3w these alloca- tions were made or what are Included in the numbers. In particular, it should be recognized that the railroad fac.lities and operations of the sub- urban services are common with those for the non-suburban services, including freight services. 'Thus the allocations may tend to be arbitrary and have little basis in fact. The report for the Central Railroad for 1966-67 was not supplied, so comparable figures are not available. Nevertheless, the following estimate was given for the loss on electric, multiple-unit ser- vices of the Central Railroad of Bombay in 1968-69.  -7- Rs. Crores 1. Total Working Expenses 7.606 2. Depreciation Charges on Rolling Stock at Current Prices and Depreciation as Booked in respect of Other Assets 1.291 3. Interest on Dividend Charges as Booked 1.311 4. Fully DiL:tributed Costs (1+2+3) 10.208 5. Gross Earnings 7.720 6. Loss 2.488 17. It is clear from these figures that gross earnings exceed what the Central Railroad calls "Total Working Expenses" which, in turn, exclude depreciation and interest charges. It would appear that total working ex- penses are greater than out-of-pocket costs andl probably are greater than ma- ginal costs. I,, any event, the figures suggest moderately sizeable losses !. the year 1968-69. 18. It appears that both the Western and Central Railroads are moder- ately profitable, although, in evaluating railroads' capital stock, it is often difficult to place appropriate values on their assets. For the year ending 'March 31, 1969, the Central Railroad reported a profit after all ex- penses of Rs. 26.2 crores. In the previous year, a profit of Rs. 39.3 crores was reported. The Western Railroad was formed in 1951 and has had a return on capital -- as they measure it -- in excess of 9 nercent in every year since its formation. In 1969-70, the Western Railroad reported a profit of Rs. 33.2 crores. If, in fact, the suburban services are operated at a loss, there are sufficient revenues in both railroads to subsidize the suburban operations as the accounting is presently reported. Hlowever, the cross- subsidies implied are terribly complex, with freight and mainline passenger service sub -dizing low-priced rail commuter service. This may have the effect of disiorting the allocation of fr-ight between rail and truck in addition to (istorting the land use allocation decisions within the Bombay metropolitan area. Studies of Future Requirements 19. Studies by the Traffic Cell have estimated the daily volumes of transit passenger trips crossing screenlines 1/ in 1968 and 1981 as shown in Table 2. These are derived from estimates of the present and future distributions of population, land uses and economic activities by using 1/ Arbitrary lines (shown on Map No. 2) running generally east-west across Bombay, dividing it into a series of strip zones.  - 8 - multiple regression analyses techiques. The estimates are for hotil rail- passenger and bus-passenger trips. Besides the usuql caveats about the accuracy of forecasts based on other forecasts whicih thenselves are of doubt- ful validity, these estimates of future transit passenger trips have some built-in procedural weaknesses. 1/ .1owever, they are the best estimates available and serve to illustrate orders-of-magnitude of expected increases in transit passenger trips. 20. The Traffic Coll studies preceded the preparation of the Metropoli- rnn Regional Plan; consequentlv, these estimates do not reflect tile popula- tion and activ1ty patterns proposed by that plan. Tlhe encire Twin-City con- cept, withl its resualtant impacts on the snnpe of regional developmcnit and travel patterns, has been excluded from consideration. Tlic esti-nate; do incltide an assumption that the Backhay Reclamation Sclhemec will be varried out and will resuitt In thle employmnellt of 150,00( additional persons In that area (whlich exp ;:.ns the unusually large jump in trips acioss Screenline No. 1). 21. Two features stand ou: in Table 2: the volumes crossing Screen- lines 3 - 6 (roughly the Bombay Central-Dadar vicinity) are substantially higher than in other areas, and traffic growth is highest in the northern suburbs (crossing Screenlines 7 and 8). 22. Using these estimates of total transit passenger trips, the Traf- fic Cell made a further subdivision into estimates of rail passengers an'j bus passengers. The split of transit passengers between these modes was based on a "minimum tice path" traffic assignment technique. Travel times between all zones were compar'!d vla the shortest (time) rail routes and bus routes, and an "all-or-nothing" assignment was made; i.e., if the travel time was shorter by rail than by bus, all trips -ere assigned to the rail system. This tech- nique has serious shortcomings. If the estimated travel time between two zones is 40 minuces by bus and 39 minutes by rail, for example, all trips are assigned to the rail _ystem and none to the bus system. 2/ The techniquc also ignores any diVference in travel costs between the two modes. The modal split is obviously hi hly sensitive to assumptions about locations of facilities and their operating speeds. Present (1968) travel times, as observed by actual measurements, were used for the road system while new rail transit facilities were assumed to have 60 kph operating speeds. I/ The estimates are for transit trips only; no comparison has been made with separate estimates of automobilp passenger trips (by Wilbur Smith) or with independent estimates of tctal trips which would be generated. The equations for estimating transit are based solely on pregently observed behavior without regard to possible variations in trip pricing, automobile ownership, or other factors. 2/ See paras. 26-27 for further discussion of this point.  - 9 - 23. Traffic assignments were made for two alter.-ative rail transit plans; the bus routes (road plan) remained the same for both. 1/ Plan A in- cluded a new rail transit corridor between the Central Railwa-. ard P.D. Hello Road, crossing the Parbour branch at Raoli Juniction, extending a].cng the Sewri Expressway at Wadala, and terminating at Ghatkopar near the Eastern Express Hiighway. Another new corridor generally follows along the west side of the Western Railway to Bandra and Santacruz Airport. These corridors would be linked between Santacruz and Ghatkopar, and between Churchgate and Victo,'d ,Terminus. Several other intermediate cross connections were proposed but after preliminary testing it was decided that bus connections would be more suitable than rail. Plan B included all of the features of Plan A plus a rail corrido; frum Kurla, running throughl the center of the island and meet- ing the western corridor near Marine Lines Station. The central-island cor- ridor was added after traffic assignments on the two-corridor system of Plan A indicated overloading of the western corridor. 24. Table No. 3 shows the daily transit passenger trip assignments to the two rail transit plans, based on the Traffic Cell's estimates of trip- making patterns for 1981. Only trips which were assigned either to the ex- isting rail lines or to the proposed rail transit lines are shown in this table; bus passenger trips are not included. 25. Recalling the sequence of the Traffic Cell's studies, Plan A was the first proposal which was tested. When the traffic assignments indicated the Eastern and Western Transit Corridors would be "tremendously overloaded," 2/ a new plan was tested which included a central corridor. Several observa- tions should be made. First, the traffic assignments are based on estimates of total daily rail passenger movements in both directions. In order to draw any conclusion about overloading, it is naecessary to convert these daily fig- ures into directional movements during peak hours, recognizing that the per- centage of total travel occurring in the peak hour a -. the directional split of that traffic will vary for each section of line. 26. A .econd, and seemingly more important criticism of these assign- ments, may b seen by comparing the daily volumes assigned to the WeFtern Transit Corridor and to the Western Railway. As was previously poir ted out, the assignm-_.s were ;ade on an "all or nothing" basis and, moreover, they were made without regard to the available capacity of each facility. The result was that Plan A, for example, had 1,552,940 daily trips assigned to the new Westera Transit Corrl.Ior where it crosses Screenline No. 5. while the existing Western Railroad was assigned only ,,9300 daily trips, or about 4 trainloads for both directions of travel. These heavy assignments of traf- fic on the proposed new transit line are principally the result of assumed 1/ Interestingly, newtther assignment plan assumed that the freeways pro- posed by the Wilbur Smitlh Study would be built or available for transit use 2/ Traffic Cell Report, p. vi-32.  - 10 - higher operating speeds (and therefore reduced travel times) than on the existing lines. The assigned volumes, in tuni, were interpreted as the jus- tification for an additional Central Transit Corridor which was tested in Plan B. The central Transit Corridor does divert 344,000 trips from the Western Transit Corridor at Screenline No. 5, but still leaves the Western Railway with only 7,900 daily trips. 1/ 27. Although the reports do not expUicitly state it in this way, the implication in finding that heavy 1981 traffic volumes "justify" the con- struction of new Western, Central, and Eastern Transit corridors is that commuter passenger traffic on thf Westerr. Railway will virtually cease and on the Central Railway lines will be cut to a fraction of the present levels. Obviously this would be inconsistent with the need to make maximum use of availabie resources and facilities before embarking on costly new investment programs for new facilities. Furthermore, it is contrary to the railroad's present studies and construction programs which are aimed at expanding the capabilities of their existing systeni. Cuirent Improvement Programs and Pt.oposals 28. The Western Railway is quadrupling its tracks in the section be- tween Churchgate and Grants Road Stations. This work, estimated to cost Rs. 4.15 crores, has been ui:Oervay since 1966 and is intended to provide additional capacity and operating flexibility in the southern portion of the island. 2/ When. completed at the end of 1972, the Western Railway will have four suburban lines extending the full distance from Churchgate to Borivli, (A later stage would extead the quadruple lines from Borivli to Virar.) The most difficult problems encountered io carrying out this work ha-e been those related to right-of-way acquisition, despite a fairly com- plete kit of administrative and legislative tools for accomplishing this. One of the knottiest problems is still un;eeolved: the detailed method for bringing .he new lines into Churchgate Station. The railroad proposes to widen the station to the east and build new platfortrs in the space now taken bv Maharishi Karve Road. This, in turn, requires that the road be shifted about half the roadway width eastward along several bleck lengths, taking nearly all of the property between the present edge of street and the build- ing line. Juidging from previoua difficulties in acquiring land under much less comnplicated circumstar'es, it would not be surprising if the track widening were completed for the remai%der of the project before the right- of-way problems are settled in this last and most crucial secticn. 1/ There are other peculia, lies in the Plan B rssignment. For example, although the Central Tre it Corridor is very near the Central Railway line, no traffic Is diverLed from the railway but r-affic is diverted from the more distant Eastern TrAnsit Corridor. 2/ There may, in fact, be =cre capacity in this section ic relation to traffic demands than on sections further north, where passerger loads are apparently much higher.  - 11 - 29. The Central Railway has made a preliminary study of the costs of modifying its lines and stations to permit the use of 12 car rakes instead of the present 8 or 9 car rakes. This study estimates that it would cost about Rs 40 crores to make this change, including the revisions in stations, tracks, and train control systems and the purchase of additional rolling stock. 1/ This investment would increase peak period' capacity on the Central Railway by about 50% over present limits. No decision to proceed with th's work has yet been made. rhe Western Railway does not consider that it is feasible to expand its rakes from 9 to 12 cars, as proposed for the Ce.tral Railway, although a detailed study apparently has not been made to de-ermine what the costs and benefits would be for a similar modification of the Western Railway lines. 30. With respect to longer range rail-transit services, there is a profusion of differing proposals. 31. At the conclusion of its lengthy analyses of travel patterns in the Bombay Metropolitan Area, the Traffic Cell report makes the following rathier cautious recommendation: "In the conte::L of the continued development in the Backbay Reclamation area on the southern tip of Bombay and expansion of suburban areas, it will be necessary to have a loop consisting of three corridors joining Santacruz, Kurla, Chembur areas to south of Bombay. It is, however, recommended for the present, to implement only one of the corridors in addition to the third terminal at Ballard Estate to relieve the existing overcrowding and for catering to the increased traffic to some extent." 2/ The "third terminal at Ballard Estate" is a reference to an earlier recom- mendation (1968) by the Metropolitan Transport Team of the Planning Commis- sion. That team concluded that the commuter problems in the Fort Area could be substantially relieved by using the lines of the Bombay Port Trust Rail- road (generally following along the east side of P.D; Mello Road) for com- muter traffic and building a new passenger terminal at Ballard Pier. This proposal was intended to add line capacity and to relieve thie passenger dis- persal problem in the Fort Area by providing the new station in an area of heavy concentration of trips. 3/ The same Metropolitan Transport Team, after reviewing t'e traffic and transportation problems of Bombay, recommended a mass rapil transit system should be provided in a north-south direction in 1/ The study itself was not available to the mission so it is not c!ear whether the RB. 40 crores include the elimination of grade level cros- sings on these lines. 2/ P. 16, Summary, Traffic Cell Report. 3/ See p2ras. 174-175 for further discussion of this proposal.  - 12 - corridors to be identified by further studies. These were the studies which subsequently were conducted by tthe Traffic Cell. 32. Map No. 2 shows the Plan b rail transit system proposed by thie 1'raffic Cell. Its main features are thle three north-south corridors, a loop connecting Churchgate and Victoria Terminuis, and a northern loop between Ghatkopar and Santacruz. In presenting this proposal, the Traffic Cell sug- gested that engineering feasibility studies of all thrce corridors should be undertaken but that only one of the corridors (unidenitified) be implemented, in eddition to the Hallard Estate Terminal, until major decisionis are reachiI about further flackhay Reclamation. Bai'dra-Kurla dc.elopment, anti the T'1iin City. l/ 33. The Hetropolitan Tranisport Team of the Planninig Commiss 'on accupted the findings of the Traffic Cell studies concerning the genierai locations of the three north-south corridors, but proposed actual rail transit alignrments which were significantly different. The Port Trust Railroad, converted to passenger commuter service, was proposed for the Eastern Transit Corridor in lieu of the Traffic Cell's alignment further to the west. The Central Cor- ridor proposal was nearly the same as suggested by the Traffic Cell except that it would terminate at Matunga, just west of Raoli Junction, rather than extending across Mahim Creek to Kurla. Also, at the southern end, it would loop around the Oval Maida:t, staying to the east of Churchgate Station rather than joining the Weszern Transit Corridor near Marine Lines. The Western Transit Corridor wou.ld be shortened considerably, veering south-eastward at a point near Lower Parel Srttion and joining the Central Corridor about op- posite Bombay Central. A separate loop was then proposed to start at Charni Road and follow the Backbay shoreline to the southern end of the reclamation area before looping back to Victoria Terminus. No traffic assignments were made for this proposed system, although the results could be expected to differ appreciably from Plan B. 34. As a result of the Metropolitan Transport Team's recommendations that further studies should be conducted, the Bombay Metropolitan Transport Project (Railways) was established in 1969 and given the responsibility to: (i) assess commuter - fic demands at present and for the next 10/15/20/25 years, (ii) study methods for optimizing commuter transport capabilities on ex,sting rail lines; (iii) make preliminary engineering feasibility studies for the Ballard Estate Terminal and connecting lines from Wadala; l/ Traffih Cell Report, pp. vii - 23. Although this report recommends con- struction of the Ballard Estate Terminal, the traffic assignment studies did not include the Port Trust Railroad line, which would serve this terminal, in their transit system so no estimate was made oi passenger traffic which would use this line and station.  - 13 - (iv) study passeng . dispersal problems in the Fort Area; and (v) make teclhno-economic feasibility studies for a Mass Rapid Transit System. 1/ 35. After reviewing the previous studies and making some rough c.alcula- tions of traffic demands and rail line capacities, the Metropolitan Tcansport Project (Railways) proposed that the following additional rail lines should be studied: (i) pair of lines, Malad-Bombay Central, (ii) pair of lines, Thana-Kurla, (iii) Jogeshwari - Churchgate corridor, (iv) Chatkopar - Victoria Terminus corridor, (v) Kurla-Bandra, (vi) northern connector between (lii) and (iv), (vii) Churchgate - Victoria Terminus locp via Backbay, (viii) Churchgate - Victoria Terminus loop via Museum and Sachivalaya. These studies would be in addition to the Ballard Estate Terminal Study (re- portedly already underway) and the studies for optimization of existing rail lines. 36. These proposals are confined to the Greater Bom'.ay area and do not consider what additional and complementary facilities are needed in the re- mainder of the metropolitan area. The proposals would add about 110 km of new rail lines to the 90 km now existing in Greater Bombay. Of these 110 km, abont 62 km are proposed to be underground and 48 km on the surface. The Metropolitan Transport Project assumed surface lines would cost Rs. 1 crore per km and underground lines would cost Rs. 6 crores per km, resulting in an estimated cost of Rs. 48 crores for surface lines and Rs. 372 crores for underground lines. The combined total of Rs. 420 crores does not appear to include the costs of additional stationis, land acquisition, or the rolling stock and other accessory equipment required to operate on these lines. 37. To complete thc picture of current proposals for rail transit im- provements, the Metropolitan Regional Planning Board has also made transit recommendations in its report "Strategy for Bombay Metropolitan Region." 2/ This report makes the following statements alout mass transit: 1/ Railway Board Lcttp- dated Jinuary 11, 1969. 2/ Published January 27, 1970.  - 14 - "(ii) Mass Tranisit. The following railway progranmmes for augmentation of commuter travel facilLties within the metro- politan area will have to be Included In thie Regional Dev.lop- ment Plan. (a) Quadruplication of Railway tracks between Churchgate- Grant Road and Borivli-Virar sections of thie Western Railway and the sections beyond Kalyan on the Central Railway is immediately reqluired. Of these, work on Churchgate-Grant Road section is already on hand. (b) Introductioni of twelve coachi tra1iis on the Central RaJlway and related improvemeuiits. (c) Near;ures for improving frequency of trains by improvements in signallinig systems, yard-remodelling, etc. (d'! The developmenit of one additional traffic corridor for mass transportation and provision of a third terminal in Ballard Estate. (e) Shifting of the mainline terminus to be between Dadar and Parel is possible. (f) Feasibility studies for the underground railway. (g) Augmentation of bus fleet by adding large capacity buses. (h) Increased 'ins depot and workshop facilities. "Otlher Urban Areas: Organization of Mass Transit ser- vices and City bus services in existing cities would be a matter of details. Adequate provision for mass transportation services will also certainly be needed in the two major town- ships fiz., Trans-Thana and Metro Centre. The goods lines proposed/existing in these areas will have to be in course of time upgraded to passenger-cum-goods lines and additional lines and stations will also be necessary." Observations on Mass Transit Studies and Proposals 38. The mission had neither the time nor the necessary information to make an exhaustive examination of the various proposals which have been put forth for rail mass transit nprovements. However, there is sufficient basis for making the following general observations which may suggest possible d!rections for further studies. (i) To state a point which is se obvious that it hardly aeeds Etating, the groups which have studied Bombay's mass transit problems have  - 15 - widely differing viewpoints about what should be done to improve the present situation and to accommodate the expected future travel den,ands. I/ (ii) None of the proposals (except for modification of the Central Railway to hancile 12-car rakes) are yet supported by studies which would give a valid prelirninary indication of their capital and operating costs, their ecosv,mic and other benefits, or their physical and financial feasibility. Altsough same useful background studies have been carried out, none of these stu.dies havs dealt specifically with the proposals which are being presented. Even the Traffic Cell's rather laborate traffic assignment studies tested systems which differ substantially from those which are now proposed. (The testing procedures, as pointed out in para. 19, had inherent weaknesses wh4cn severely limited their usefulness even for the systems which were tested.) (iii) .'ll of the proposals involve high financial costs and must therefore be considered in the context of total resource availability and other competing demands. The initial :ost of the Metropolitan Transport Project proposals (roughly estimated at Rs 420 crores), for example, would impose a severe burden on the total resources of the metropolitan region, regardless of the "need" that might be demonstrated for these investments. (iv) Studies which have been conducted to date do not refLect cur- rent proposals for the future development of the Bombay Metropolitan Region, as described by the draft plan and report of the Metropolitan Regional Planning Board. Conversely, the Metropolitan Regional Plan has not been examined in the light of its implications for future transport requirements. As a result, the mass transit proposals have not been integrated, eatier in area of cover- age or substance, with the proposals for metropolitan regional development. There is at least a reasonable possibility that the two sets of proposals may be working at cross-purposes. 39. The Metropolitan Transport Project (Rallwa-s) has clearly indicated its recognition that additional studies are needed to provide an adequate basis for decisions about future mass transit projects. It has recommended a specific list of studies 2/ which it considers should be carried out and has suggested that all of these should be undertaken simulta:neously. The studies are identified by location only; the precise scope and nature of the proposed studies are not described. 40. ihe notion of simultaneous studies should be carried one step fur- ther. What Is needed is a coordinated study program which begins with the consideration of the entire t:ansport system, including both highways and 1/ It might also be pointed out that there has been a secies of earlier studies, not mentioned here, whichi have made recommendations differing from all of the present proposals and from each other. They are all so out of date that they have not been included in this discussion. 2/ See para. 35.  - 16 - mass transit facilities, an( the plans for regional development. After ex- amining various system a.ternatives witlhin this physical, social and econon,ic planning framework (observing reasonable order-of-magnitude constraints on investment3), the details and timinig of the comnno.ent projects can thien be determined. Assurance that the be;t investment decisions are being riade, given the present and forecast future circumstanlces and the constraints on resources, can only be gained by Looking first at the system and thien at the projects. The reverse procedure, starting witlh detailed examinatiolns of projects, does noc yield the same results. 41. When one considers thte size and complexity of transport analyses, as well as the importance of updatinig decisions to reflect changing condi- tions, it becomes apparent thac a continuinig plannin$ program is re(lUirLd, not simply an ad hoc study. Moreover, the transit planning activirv slhould be carried on by, or at least have substantial participation of, tLle agency or agencies which ultimately have the responsibility for building and operat- ing the system. At least during the early stages and for highly specialized tasks throughout the plcnning program, consultants' services will undoubtedly be required; however, a concerted effort should be made to develop ir-house staff capabilities for the major portion of the planning activity. Alternatives so Consider 42. Wthen identifying alternative courses of action to consider, it is important to know the desired objectives and, equally important, to eliminate any artificial constraints which might cause worthwhile possibilities to be overlooked. This is parEicularly true in the case of decisions for urban transport investments, wbere the goal is not simply to move more people. Tthe best urban transport "solution" may be one in which new investments in trans- port services are a relatively minor part. For example, the mass transit projects which have been proposed for Bombay are aimed at improving the quality of present services (by reducing crowded conditions, improving re- liability and punctuality, etc.) and at providing for anticipated future growth. The present proposals would try to accomplish this by providing additional rail lines, more trains, bigger stations, etc. at considerable capital and operating cost. A non-transit solution to the same problem which might be equally effective and less costly would include the reloca- tion of job sites from the r!':.c Area to the northern suburbs in order to re- duce the number of persons entering the Fort Area, to shorten the trip for persons now living in the suburbs, and to permit commuters from the Fort Area to use vacant space on existing outbound trains. (Some of the railroad's own 90,000 employees might be a good starting point for this kind of movementl) This approach has the advantage of easing the problem on three fronts--re- ducing overa.l travol, reducing passenger dispersion difficulties in the Fort neea, and making more efficient use of present facilities--while pro- viding additional capacity is certain to worsen the passenger dispersal situation and the unbalanced directional movement. 43. Another alternative which appears to merit serioue study would be to move all main-line passenger services from Victoria Terminus and Church- gate to a new terminal, possibly in the Bandra-Kurla or Mahim-Sion vicinity.  - 17 - A new rail passenger terminal in either area could be part of a complete transportation complex, with connectiag services to suburban commuter lines, metropolitan buses, taxis, airport shuttles, etc. A transportation terminal of this type would also be a focal point for stimulating other commercial and office development. The main advantages from a transportation viewpoint would be to free the existing riglhts-of-way and stations in the island fo- exclusive use by metropolitan conmuter trains during peak traffic periods. 1/ This would permit track layouts to be modified for more efficient operations and would permit complete flexibility of track scheduling. 2/ The current proposals by the Metropolitan Transport Project are based on assumptions that no commuter trains will be permitted to use main-line tracks by 1981. This assumption should be carefully reexamined in terms of the trade-offs which are involved, considering that daily passengers on the main line trains number in the thousands while the daily commuter passengers already exceed 2 million. There is no evidence to indicate that the main-line passengers would be inconvenienced by moving to a new metropolitan area terminal, given the present patterns of development in Bombay and the wide range of connect- ing services which could be provided at a new l1cation further north. 44. A new metropolitan transportation center could also provide better interconnection between both the main-line and the commuter passenger ser- vices of the Central and the Western Railways. This capability for inter- changing between lines will become increasingly important as new growth ex- tends further to the north and eastward across Thana Creek. Some capability exists for inter-line transfers at Dadar and Mahira-Raoli Junction but the transfer arrangements are inconvenient and probably inadequate to cope wit: growing demands. In any event, the full capability for suburban-mainline- interline passenger exchange does not now exist. 45. One feature that has been included in several of the mass transit proposals is the construction of a connecting rail loop between Churchgate and Victoria Terminus. Among the possible advantages of this loop would be: (i) the dispersion of some passengers who now use Churchgate and Vistoria Terminus to one or two new stations nearer their places of work; and, 1/ Freight traffic also uses these lines but could be moved duriag off peak hours. Tnere also appear to be opportunities for longer range modifications in freight handling facilities and operations which would alter the track-use requirements in Bombay. (See para. 178.) 2/ In planning and carrying out all track improvements, including those now underway, possibilities for future changes in opcrations should be kept in mind. There .may be opportunilties for Improved operations whlich could be preserved or enhanced at relatively nominial costs at thc time of construction but which would be muclh more cobtly to pro- vide by revistons at a later date.  - 18 - (ii) the elimination of dead-end train turnaroundIs at both stations, thereby pernmitting closcr headways and im- proved operations. For the full realization of either of these benefits, the loop would have to be an extension of the incoming tracks at bothi stations. This would be very costly, regardless of how it might be accomplished. Existing buildings and cross-streets would make it difficulL, if not completely impractical, to con- strtict the loop at grour,d level to meet existing tracks. If the loop were constructe(d underground, the present tracks should also go underground be- fore reachinlg both stations (a costly solution, but one that could have some attracrt'e features). A separate underground line would require passenger transfers at Churclhgate and Victoria Terminus and would not permit the straiglht-through train operations. Detailed studies would be required to compare the potential benefits and costs of the loop. The number of pas- sengers who would use this loop is quite small in comparison to the rider- ship on existing sections of either railway (and also hiighly dependent on whether the Backbay Reclamation continues or not). Provision of a loop be- tween the Central and the Western lines would certainly make it essential to reconsider the entire question of separate or unified passenger trans- portation operations in the metropolitan area. Improvements of Existing Services 46. Regardless of whatever shortcomings they now have, it is clear that the present railroads must continue to bear the main burden of commute. traf- fic in Bombay for some time to come. If it were possible to decide immediate.- 'y that a new and separate rail transit system should be built (and ;here is not yet an adequate basis for making that decision), seven to ten years can easily el.pse before such a system would be in operation. Further studies are needed of system configuration, preliminary and detailed engineering de- signs must be prepared, financial plans and schedules must be worked out, rights-of-way acquired, etc. before actuzl construction could begin. In short, if the mass transit passenger traffi.. which has been forecast for 1981 actually develops, the probability is high ftat these passengers will be riding on existing rail and bus systems. 47. There are at least two clear conclusions to draw from this. First, the continuing program of metropolitan area transport and planning studies which is required to guide investment decisions should be initiated aj soon as possible. Second, the highest priority siiould be given to the studies and resulting investments which will permit better utilization of the existing rail systems. In fact, only after the optimum capabiliti2s of the existing system have been identified will it be possible to assess the need for build- ing new systems. 48. The "optimization" studies shouli consider all possibilities for improving present services. Some of the more obvious ways include the fol- lewing:  - 19 - (i) Equipment modernization. The 1928 rakes still being used by the Western Railway carried 20% fewer passengers than th- 1964 rakes 1/ and even the 1952 rakes carry 140 fewer passengers than the 1964 rakes. Replacement of old cars would increase train capacities and also improve train operations, besides reducing maintenance and repairs. Serious consideration should be given to modifying car interiors, possibly removing space-wasting bulkheads and revising the seating arrangements. By removing some of the seats, at least in trains operating on shorter runs, capacity could be increased substantially. (ii) Longer trains. The preliminary findings of the Central Railway regarding the use of 12-car rakes slhould be exam- ined in more detail and similar studies should be mad- for the Western Railway. Even changing to 10-car rakes would increase the capacity of each train by 11%. Before making a definite decision to change to 12 car rakes, otlher pos- sibilities should be examined to determine which is the nost economic size, in terms of costs and benefits. (iii) Level-grade crossing elimination. Train delays would be reduced and safety and punctuality 'mproved if separated crossings were provided at major roads and if more effective crossing protection were provided at minor streets. Road traffic would also benefit from the uninterrupted flow at separated crossings. (iv) "Bottleneck" elimination. Sharp curves and other track layout features restrict operating speeds on some sections and add to the turnaround time at stations. Even if the correction of these conditions requires additional right of way, tl.e benefits may still justify this. (v) Operational Revisions. Train scheduling and operating procedures should be examined to assure that the most appropriate and efficient techniques ore being used, in- cluJing shuttling, station-skippinp, express trains, and different allocations of tracks fpr commuter and mainline operations. 49. The foregoing suggestions all relate to possibilitieb for improving line capacity. Most of the measures wlich might ibiprove the collection and dispersion of passengers in the vicinity of Churchgate and Victoria Terminus are not the railroad's responsibilities and require actions by other agencies. Some of the possibilities that are worth investigating Include: 1/ Base(d on Traffic Cell's calculations of seated capacity plus standing capacity, all;.wing 1.3 sq. ft. per standing passenger.  - 20 - (i) Bus service improvements. Frequent shuttle buses, operat- ling from off-street loading zones near the railroad stations directly to a few key locations in the Fort Area and Colaba which are beyond a convenient walking distance, could ease congestion around the stations and provide a very useful service. (ii) Intersecti)n traffic control improvements. Modern traffic control :quipment and improved intersection designs, coupled with better regulatic' of pedestrians, at nalr, strec: Inter- sections near the stationis would permit faster and safer move- ments of vehicles and pedestrians. (iii) Separated pedestrian crossings. There may be justifica'on for conLtructing separated pedestrian crossings, leading directly from the stations and passing either over o; under the adjacent streets, in order to remove the pedestrian- vehicle interference completely when both flows are very high. 50. It should be emphasized that these suggestions for better utiliza- tion of existing facilities, for the most part, are neither new nor startling. Similar suggestions have been made by t.he Metropolitan Regional Planning Board, the Metropolitan Transport Team and the Traffic Cell, and others. For example, engineers who studied the operations of the Western Railway for the Juhu Chamber of Commerce concluded that tbe speed and frequency of suburban commuter services could be improved significantly, simply by revising sched- uling and operating procedures fcr existing equipment and facilities. Gener- ally, however, these suggestions havz not been followed by the necessary de- tailed studies which would ana4yze their feasibility in terms of costs and benefits and would provide a basis for decisions and improvement programs.  - 21 - II. RAIL GOODS TRAFFIC 51. The total broad ga;ige rail car interchanges between the Western and Central railways are very substantial. The Central Railway in the fis- cal year 1968/69 received 1,732 wagons per day from the Western Railway, in- zluding 1,031 loaded and 701 empty. The average daily dispatches from the Centre' Railway to the Westein Railway amounted to approximately the same number of cars, 1,750 of which 1,365 were loaded and 385 were empty. Al- though the n.zbers seem extremely large, there is fair agreement between the Central Railway report and the Western Railway report for the year 1968/69. Of this very large number of interchanges between the two princi- pal railroads serving Bombay, app;oximately 400 are made in the metropoli- tan area of Bombay, at Dadar Junction and at the Bombay Port Trust at Wadala. 52. In 1969/70, 196 cars were dispatched f'r.' Dadar Junction including 179 loaded and 17 empty, and 242 were dispatched from the Wadala Terminal, 157 loaded and 85 empty. In the same year, the average daily receipts were 220 cars (208 loaded and 12 empty) at Dadar Junction and 240 cars per day (168 loaded and 72 empty) at Wadala. This interchange at each of the sta- tions Dadar and Wadala amounts to approximately the equivalent of 4 trains at each station in each direction for the Western Railroad. In 1969/70 the Western Railroad handled 959 million wagon-kilometers and 16,193,000 goods train-kilometers or a weighted average of about 59 wagont-kilometers per train-kilomeZer, suggesting a train length of slightly under 60 wagons in the broad gauge operation of the Western Railway. 53. Although this interchange does not represent a very large propor- tion of the total interchanges made by either the Central Railway or the Western Railway, the volume of interchange taking place in the already very congested railway operations in the Bombay metropolitan area may make for an intolerable situation. The railroads have suggested, and indeed plan to construct, a new link in the northern, less dense part of the metropolitan area to avoid congestion in Bombay City. As of the 1969/70 reports of the two railroads, however, the project has not yet been surveyed and no cost estimate is given. 54. In addition to the congestion within the Bombay metropolitan area, a major problem for the railroads is the inadequate track capacity through the Chats. There a-e presently two Ghat lines that are utilized to nearly full capacity. The capacity limitat,on is principally the result of opera- ting procedures which require large separations between trains to prevent accidents that may be caused by runaway trains. There are proposals for adding a third Chat line, but no detailed proposal or justification of the project has been developed. It would appear that since such a large portion of the movement of freight over the Ghats is petroleum and petroleum pro- ducts, it wo,.d be worthwhwle to consider as an alternative the construction Or a pipeline rather than a third Ghat line. In 1968/69, the Central Rail- way originated 716,000 tona of mineral oils, nearly all of which was bulk. In the same year, thiey terminated 928,000 tons, again principally all bulk.  - 22 - All of this traffic--except for a *ery irsignificanP. amount--MDovod on the broad gauge sections of the Central Railroad, bhLt no information is avail- able on the amount moving over the Chats. 55. Before expending large amounts of money on tihe develo) - new Chat liite, it would be appropriate to consider a large number u. :-her alternatives, including the provision of devices to stop runaway trains in the event of brake failure such as in-track braking and the provision of extra motive power so that longer trains may be moved over the Chats. 56. The probltnm of the Ghat line may Indirectly affect the BomL-'y area, since it eliminates some of tlie flexibility in scheduling that cttuld assist in relieving tne main lines In the metropolitan area during tl;e p.ak hours of commuter traffic. Also, because of scl:eduling, the storage of trains and freight is longer than might be necess v if the capacit, prohle:t did not exist. The extent of these problems is no. entirely clear; h0wever, it would seem desirable to remove any unnecessary freight problems from ctle already overburdened rail syste.w in Bombay. 57. In-addition to the storage and yard facilities located in Bombay City, the two railroads have exLt.nsive workshops in the City. These are principally at Parel. The Central Railway workshops at Parel perform all of: the work required on the Central Railway locomotives, both broad gauge and narrow gauge. Carriage and wagon work is done elsewhere. The Central Railway also performs some special repairs at their Parel workshop for the Western Railway. The Western Railway has extensive workshops for carriage renaitfsat Parel. Parel is the only broad gauge carriage workshop of the Western Railway. Thus the possibilities of moving these activiPies out of Bombay City appear remote in the short run. The difficulties of moving the extensive facilities neededl to repair locomotives and carr:ages would prob- ably preclude any immediate change. 58. The third Bombay railway, that of the Bombay Port Trust, has scme entirely different problems that are related to serving ships at dockside. Rail- cars frequently are not available when they arc needed for unloading ships-,. and material beir.g brought in to be loaded onto the ships must be b!ought to the port earlier to assure its availability when the ship arrives. Compounding the congestion at the port is the presence of rail cars blocking the access to the ships. The port consultants have suggested that the rail- way be abandoned and a road be provided to replace it (see para. 174). if the movements of cargo fron the port and to the port originate and terminate outside the Bombay metropolitan area, the use of truck transnortation rather than ra1W directly from the ships does not seem entirely satisfactory. Long-distance hauls would require a truck movement from the docks to an intermediate warehouse or .ransfer point where goods i,uld be transferred to a railroad car and then moved by rail to the destination area. Such double handling Is very expensive. It would seem that a major attempt a_ coordinating the rail operations so that the demands of ship loading ar. unloading could be met would be most appropriate. In this way double handling of import and export merchandise would not be required.  - 23 - 59. Part of the problem of sort movement results from difficulties iu. clearing Customs, both inbound and outbound. It would seem that effective coordination between the Customs operations and the transportation opera- Ltons of the port would provide greatly improved transportation and handling of freight through the Bombay port. It might also relieve some of the prob- lems presently creating a demand for the new port at Nhava Sheva. To the ezn.t that loading and rnloading operatious at the port would be expedited so that ships could be urned around more quickly, so:ne of the need for the new port might be alle,iated.  - 24 - III. BUS TRANSPORTATION IN BOMBAY The Operating Company 60. All internal bus transportation in Greater Bombay is provided by the Bombay Electric Supply and Transport Undertaking of the Municipal Corpo- raEi'n of Greater Bombay (BES&T), except for some verY limited privately operated bus services to a few factories and other special places. Outside Greater Bombay, bus services are provided by the Maharashtra State Road Transport Corporation (a nationalized corporation), which also provides intercity service and ser;ice between Greater Bombay and other areas. Thus, although Maharashtra State Road Transport Corporation operates in Bombay, it does not carry any internal local traffic. 1/ Route Structure and Schedules 61. Th- BES&T operates slightly less than 120 bus routes consisting of approximately 1,360 kil.tnmeters. This sytom is divided into 10 depot areas for which traffic and roiute information _.re reported. The route structure provides very extensive coverage of both _he city and the suburban areas of Bombay that are w-ithin the Bombay municipal limits. 62. Although buses operate on the comparatively few existing east-west roads, the bulk of the routes In the suburban area tend to -in north and south as do the railroads. Indeed, many of tha bus routes ari nearly paral- lel to the rail commu~ter system. Such greater and more ubiquitous coverage is available in the city, although there is somep fairly extensive parallel- ing of railroad operations. One would expect the bus system to provide ex- tepsive feeder st-vice to the baaically north-south oriented rai! commuter service. While this is probably the case, the route system does not have great focus on the railroad station pattern, although all stat'-ns Ere served by the bus system. There does appear to be a closer relationship betnwen reiilroad stations and bus cperations in the suburban area than in the city. 63. If the buses serv3 as important feeders and diatributors of rail passenigers, it is surprist'ag that very few bus routes orig'oate at the rail station, particularly Churchgate and Victoria Terminus. !.arge numbers of passengers embarking and disembarking from buses at these princlpal terminals create long delays to scheuules, and buses cannot generally be availrble to accumulate passengers as they l1ave the terminal. Since buses generally wait at their point of origin, a logical place to wait and provide a service while doing so is at the main ra&'raad stations. 1/ The geographic area served by the ele tric supply operatio;n of the BES&T is confined to the city of Bo3)a;y, which may cause cross subsidies to arise between transportatiorn and electric supply operations. Cross subsidation within the bus transport system is discussed below.  - 25 - 64. The BES6T provides excellent route maps of its system with de- tailed routing described. Separate maps and descriptions are provided for city and suburban services. In addition, the BES&T makes a complete fare table available which gives for each route the time of the first bus, the last bus, and the headways in addition to the fares between any points along the route. According to those tables, the shortest headways are 4 minutes and there are very few that are longer than 8 minutes. Thus, it would appear that present schedules do not reach saturation on individual rou.tes. It is difficult to say, hovever, whether individual stations or stops serving large numbers of routes are oversaturated owing to the confluence of a large number of longer headways. While congestion at station!. can be alleviated somewhat through the scattering of bus stops, such an answer is not entirely satisfactory because it would inconvenience passengers requiring a transfer. T'he Fleet 65. The transport operations of the BES&T are provided tlrougi. conven- tional and trolley buses. l/ The' bus fleet totaled 1,291 at thi.' end of August 1970, divided into 714 single-deck and 577 double-deck buses. Of the latter, 50 were articulated trailer double-deck buses. The fleet composition is even more variable. There are buses of at least three manufacturers; there are 8 cifferent model single-deck buses in the fleet and a total of 29 different types when seating configuration is considered. A large number of seating/ model combinations appear in the fleet unly two or three times. There are 26 se.ting/nodel combinations which appear in the fleet five or fewer times! This abundance of model types and seating configurations probably results in some uneconomic maintenance and operating practices; it suggests the need for larger inventories of spare parts and a larger number of maintenance prac- tices and rules to handle the large variety of buses in the f'.eet. It is difficult, however, sz know exactly what the cost of such hih' variety might be or to recommend a procedure to correct it without an infu: of subst:an- tial new capital thiat could quickly make the fleet mcre uniform. If equip- menr Is added and replaced gradually, thiere are bound to be changes in models and manufacturers, but the abundance in Bombay is far greater than one might reasonably expect. 66. Most of t:ie single-deck buses have total seating and standing ca- pacities of 50 passengers. The bulk of tlhcse have approximately 40 seats and roum for approximately 10 standees. Of the single-deck buses, however, 18 have 45 or 46 seats and room for approximately 35 standees. The range of capacity for thie double-deck buses is approximately 80 to 90. Thle articu- lated trailer buses have a capacity of approximately 100, almost entirely seated. The remainder of the double-deck buses also have mo3t of their ca- pacity In seats, with more, but still relatively few standees -- approxi.tLely 1/ T'rolley buscs are relatively in3ig1nificant In the total transport opera- tion. In the 1969/70 fiscal year. tlherc were on ly 12 trolley buses operated by the BES&T, and the averagc number of such buses in service was ll.  - '6 - 11 or 12. The capacities of the sin,;le-deck buwses withi large nunribcrs of standees are nearly equivalent to t1le smaller double-dvck buses butL have approximately 20 percenL less capacity than the articulated trailers. (f course these "standce" buses do not provi(de equivalent service in turins of seating. 67. Detailed data was not provided on the age distilbutioil of the fleet. The outward appearance of the fleet is thiat of very old equipment. This may be somewhat deceiving, towever, because double-deck buses May te:nd to look ancient even if they are currently produced. Furthermore, It ap- peared that maintenance efforts are directed more toward the mechanical running of the buses thani to their phiysicai appearance. Thc buses appear to be somewhat banged up, and body repair work Is poor. lhis may be a re- ~'lection of the BES&T's priority o0. the contitued operatlon of Ltle bus ov_r * s outward appearance. Certainly the BESET's fleet utilization of .ti.t jO percent must be regarded as quite good when the problems of sparc part:. are considered. The utilization figure mLy be somewhat distorted as the effective fleet excludes buses undergoing major overhaul. According to thle statistics, however, never more thatn 3 percent of the fleet is being over- hauled at one time. Road Traffic 68. Any public transport system utilizing public roads must operate in an environment with many other vehicles. In Bombay, the problem is compound- ed very substantially by the presence on the roads of many pedestri_ns, hand- carts, oxcarts, horsecarts, stray oxen, and other vehicles, principaily taxis and trucks. Insufficient atten.ion has been paid in Bombay to adequate traf- fic control and vehicular discipline on the roadway. Very few traffic lights exist, and in many places roundabouts or traffic circles are used in place of appropriate signalization. While at the present time traffic is not as severe as it is in many other c'ties, the poor utilization of roadway capac- ity does result in substantial congestion and dangerous practices. Pedes- trians utilize the streets as sidewalks, cross the roadway apparently at whim, with the result that traffic is considerably slowed down. This, of course, has substantial repercussions on bus operating performance and economics. If the buses could travel at higher speeds, they could be uti- lized nore effectively and produce higher revenues per day. 69. A program of strict enforcement, keeping pedestrians on the side- walk and permitting crossing only at specified intersections unde. appro- priately signaled conditions, may produce a significant improvemert in road- way operations in Bombay. Such enforcement, however, would be an enormous task. In particular, the sidewalks are presently occupied by vendors and sidewalk residents who restrict the flow of pedestrian traffic. These "zero" rent activities on the sidewalk provide a high penalty for transpor- tation on the roadway. 70. The presence of other non-motor vehiculdr traffic, such as animal- drawn carts, handcarts, and stray animals, appears to be a serious problem in only a few areas of Bombay. Nevertheless, they have a substantial impact  - 27 - on traffic flow whien and where they occur. Before embarking on a substan- tial construction program for new roads, more consideration should be given to improving the traffic flow on existing roads and streets. Througlh con- trolling traffic, substantial itncreases in road capacity may be achievable at far lower expense than building new facilities. Furthermore, the added discipline required and produced would increase safety and make *ravel more comfortable. Buses and taxis, wh,ich have to stop suddenly to avoid hitting pedestrians or animals, cannot oroduce a comfortable ride. Furthermore, sudden weaving and turning to a%oid such obstacles also diminishes quality of service. At some time in the future, it will be necessary to improve traffic conditions in Bombay; it would seem that the sooner this is done, the better. Such improvements may have far-reaching effects on the accuracy and operating efficicncy of thc present BESET bus system. Capacity and Growth 71. Data are not given in the BES&T's repor.s for capacity utilization. Figures are given for passengers carried and effective kilometers but are not broken down as to bus type, nor are comparisons made between passe :ger kilo- meters and available seat kilometers. Such statistics would be very useful in evaluating the performance of the system with respect to passenger ser- vice. Furthermore, it may suggest differences between bus types that are not apparent from physical measurements. The absence of detailed load factor information makes it difficult to determine whether there are constraints on supply which are restricting the growth in passenger transportation. In the simplest case, of course, if all buses were filled all of the time and there was no increase in either the number of buses or the seats per bus, one could not expect any growth in traffic. On the other hand, if traffic grew while capacity was well in excess of traffic at all times of day, one could argue that such growth is due to increasing demand. 72. Because of the way passenger information is obtained, it is diffi- cult to obtain the true number of passengers using the system. This is un- fortunate as such statistics could be very valuable in planning and managing the system. The principal reason for losing that information is that BES&T's data comes from single ticket srles, whiich do not always represent single trips. For example, some very long trips require two or more tickets. While there probably are not very many such trips on the bus system, there are some and they may be relatively important. In addition. data are not pre- sented for those passengers traveling free, such as policemen. Fare Structure 73. A zone system of fares is used by the BlS&T. The minimum charge for adults is 10 paise and tickets go up to 50 palise. For adult fares ex- ceeding 50 paise, a 45 paise ticket and an additional ticket are sold, making the total the amount of the fare. Children travel at half fare with a min- lmum of 5 paise. As a result of the method of charging fares in excess of 50 paise, ticket saic dAta for 5 paise tickets represent exclusively chil- dren's tickets in excess of 5 paise and tickets to make up fares in excess of 45 paise appear as lower price tickets, and it is impossible to distinguish between them.  - 28 - 74. The fare structure has a fairly sharp taper for both ordinary services and limited and express services. For ordinary services, the taper runs from 5 paise/kilometer for 2-kilometer trips to 2.12 paise/kilometer for 26-kilometer trips. On limited and express services, the taper runs from 7.5 paise/kilometer fLr 2-kilometer trips down to 2.5 paise/kilometer for 26-kilomeLer trips. 75. The orcinary fare structure was cha.aged in 1966. For ordinary services, 5 paise were added to each previous fare, and an additional cate- gory of 10 paise for a 2-kilometer trip was introduced. For the limited aud express services, the basic structure itself was changed, with a slharply increased taper. There has been essentially no change in the money fare since 1966 except for the introduction of a 35-paise fare on December 1, 1968, for limited and express service. To the extent that there has been any change in service quality, however, an implicit fare change may indeed have taken place. Such would be the case, for example, if headways have increased or if capacity were shifted from seated to standing or if fewer buses were operated. In money terms, however, it is interesting that bus fares substantially exceed third-class seasonal railroad fares. For example, a 26-kilometer bus trip would cost 65 paise for a limited and express ser- vice bus, while a third-class monthly ticket for a 27-kilometer railroad trip would cost Rs. 4.25, an amount that would be paid for 14 bus trips or 7 roundtrips. For most passengers making work trips, a monthly season ticket would probably provide for approximately 25 round trips. 76. It is not clear whether cost differences explain all of the price differences. It is difficult to determine this because for both the rail- roads and the bus system some common costs exist between the local cormuting operation and other activities of the organization providing the service. For the railroads, in addition to the commuter service, there is mainline passenger service and freight services. Much equipment and investment and, particularly, many personnel are commonly used between the commuting opera- tions and these other railroad services. A similar situation exists in the BES&T between the electric power services and the bus services. The problem is probably not sc severe in the latter case because the services are rela- tively distinct in the equipment and personnel used to provide them. Never- theless, data are not sufficiently detailed to indicate the true financial picture. It is, of course, highly likely that bus transportation is more expensive per seat mile than rail. It is just as clear, however, that there is probably a deficit in both op'eratior. given the existing fare structures. Facusing on bus finances for the moment according to the BES&T financial accounting, during the year 1969/70, the deficit from bus operations -- even if the total of general administration allocated and the depreciation charges were subject to question and 'eli-minated from the calculations -- would still be present. 1/ Such is not the case, h6ivever, with respect to the trolley bus 1/ It is impossible to evaluatee the reciation figures as presented with- out a great deal more detail r^ga ..tg tthe assets being depreciated. For economic analysis purpo