Eric Bruun, 

Systems Consulting Group, LLC

Baltimore, MD

Edward Morlok, 

Department of Systems Engineering

University of Pennsylvania

293 Towne Building

Philadelphia, PA 19104-6315

Francis Vanek*

Consultant

Ithaca, New York

*Corresponding author.Email: francisvanek@yahoo.com

July 30, 2001

Word count: 4,052 + 4 tables, 2 figures

ABSTRACT

Introduction 

In the present day in the United States, conventional bus transportation systems are unable to provide a viable alternative for many types of urban trips, as underscored by their very small market share in almost all cities and regions. One market that has grown rapidly in the past several decades but which is poorly served by conventional bus transit is intra-suburban travel. While transit continues to serve many trips to the Central Business District, it serves only a small portion of the rapidly growing number of trips, both for work and for other purposes, in suburban and other low density areas. 

Expanded use of minibuses in place of full size buses can help fill the need to serve these low-density markets.This minibus service is envisioned to include a number of features: 1) ability to penetrate suburban road networks, 2) ability to serve wheelchair bound travelers and others who otherwise require a separate paratransit service, 3) a high level of service and upscale image similar to that of commuter rail service or modern heavy rail systems such as those of Washington, Atlanta, or the Bay Area.

Additionally, the use of Intelligent Transportation Systems (ITS) technology to monitor system performance can further enhance the reliability and hence appeal of the system to riders.These systems can be used both to monitor the location and punctuality of vehicles in real time and convey this information to riders in a form which is easy to obtain and understand.This ITS application has been further developed by the authors under the title of “The Advanced Minibus Concept” elsewhere (1).

Public transport agencies across the US are already discovering the appeal of the minibus for a wide range of urban transportation applications.For example, a temporary circulator to link a shopping mall to the Washington Metro during a disruption from freeway construction is being made permanent (2).In the Fort Lauderdale area, a large number of persons with disabilities or without autos depend on transit. Community shuttles are so popular, that half of the revenues from a recent tax increase to expand transit went to minibus service (3). The transit agency serving the Philadelphia suburbs began its first minibus service in 1996, and then added another in 1997. In 1998, after success with its first two, it bought 50 minibuses more to equip a total of 6 routes (4).When the Miami busway extension to the rapid transit line was built, it was equipped with 46 minibuses that first circulate within communities and then use the buswayto the rapid transit terminal (5).These examples likely represent just a small subset of the many situations where agencies have opted for the smaller size and mobility of these vehicles.

In this paper we will first discuss the advantages of using minibuses in general terms.We then focus on three specific analytical techniques to compare regular bus to minibus services requiring approximately the same operating budget. For the first two, we use a case study based on a route within the Southeastern Pennsylvania Transportation Authority system that is a likely candidate for conversion. For the third we create a hypothetical service.Finally, we discuss our findings and the prospects for greater use of minibus service.

The Minibus Service Concept 

Starting with those situations where the standard fixed route version of the concept is applicable, the buses would operate on a schedule that is very frequent by conventional suburban transit standards--15 minutes or less. This minimizes waiting time for those who choose to arrive at random at stops, and for the short distance market to which it is aimed, ensures a short overall travel time. Comfort is also critical, and both vehicle and stops would be designed with this in mind. Buses could have comfortable cushioned seats, be fully sound-proofed, and have background music, much like the vehicles used by rental companies at airports. Drivers could be trained in courtesy and dealing with the public. Shelters are provided at stops, and, because of the small size of the bus, these can be located away from traffic if necessary, to enhance the quality of the stop environment.

Buses and stops could be fully equipped for accommodating wheelchair bound travelers and others with infirmities or disabilities. This would include a low floor with wheelchair ramp on the bus, and carefully constructed stop pavements so that the proper alignment of ramp and sidewalk is easily achieved by the driver. In addition, the small size of the vehicle and its low floor will make it easier for passengers with infirmities, such as many elderly citizens, to use the service. The proximity of the driver should make all travelers with special needs more comfortable.With modest and prudent slack in the schedule (i.e., a cushion of time to permit one or two special stops), this should enable either accommodating a single wheelchair-bound traveler without devastating the schedule--a serious problem with larger high floor buses where the delays for boarding and alighting extend to many minutes. 

In instances where minibuses are introduced in conjunction with an advanced communication and control system, (Intelligent Transportation System technologies), passengers can also request a deviation from the route in order to arrive directly at their destination. This deviation for boarding passengers can be controlled by either the dispatcher or driver such that they are made only when adequate time or slack exists in the schedule. Passengers calling for such a deviation in order to board can be told via whatever communication medium is used that the next bus will not be able to stop, but that the following one, expected at a specified time, will do so. This concept has already been successfully developed at Potomac and Rappahannock Transportation Commission. It has indeed also addressed the needs of some persons with infirmities and disabilities, while also sparing the cost of a separate paratransit service(6).

A comparison of the general features of high-quality minibus (MB) service to regular bus (RB) service is given in Table 1. 

{Table 1 about here}

Breakeven Comparisons of Regular Bus and Minibus Services

In the last few years, highly attractive new models of small and medium size buses have become available in North America, such that there are now vehicles for every price range and life span. There is a wide range of features available for buses of all sizes, with a wide range of interior layouts and capacities. To investigate the effect of each variable would not be possible in a paper of this length, and might only serve to blur the basic points of the article. Thus, the number of variables is minimized to those that are most crucial for showing the tradeoffs of minibus versus regular bus. These include the relative operating cost and the level of demand. Other variables, such as number of spaces for passengers and the desired load factor, are chosen as typical values in use at many agencies.

Services will be compared three ways. The first assumes that a minimum level of service is to be provided on a route, and then solves for the ridership level where costs breakeven. The second assumes that ridership is constant and then solves for the frequency of service that can be offered with the same operating budget. The third investigates a hypothetical feeder service in order to show how not just service frequency, but area coverage as well, can be increased with substitution by minibuses.

The first two use the same case study, one similar to the Southeastern Pennsylvania Transportation Authority’s Route 105. It was chosen for its properties that indicate it might be a good candidate for conversion. It is a long route at 16 miles, and operates at one-hour headways all day long, although part of the route has some additional service during peak periods. The operating speed assumed is the system average of 10.2 mph, found by dividing the Operating Expense per Revenue Vehicle Hour of $92.58 by Operating Expense per Revenue Vehicle Mile of $9.08. (7). The operating speed of minibuses is assumed to be 25 percent greater than for a regular bus, typical experience from conversions in the UK in the early days of deregulation. (8)(For further details regarding the Route 105 case study see (9)).

The operating cost of a minibus relative to a regular bus can vary depending on the wage differential, the model and age of the regular bus, and the size and model of the minibus. Thus, a range of values from 50 percent to 80 percent will be computed. The assumed characteristics are summarized in Table 2.

{Table 2 about here}

Quantitative basis for breakeven analysis

The fundamental mathematical relationship used is that the hourly operating cost be the same for either the Regular Bus (RB) or MiniBus (MB) alternative. This hourly cost is the number of vehicles assigned, N, multiplied by the hourly operating cost, OC. But the integer constraint on the number of vehicles assigned means that the operating cost is never really the same. Instead the fleet size must be rounded up or down and the cost is only approximately the same. Thus, the relationship is:

The relationship to determine fleet size required when headway is specified is:

,

where T is the round-trip cycle time, v is the operating speed, L is the route length, and h is the headway. The relationship to determine fleet size when D, passenger demand per hour, is specified, is:

,

where a is the desired load factor and C is the number of passenger spaces on the bus.

Total cost including the ownership cost of the vehicle could be used instead of the operating cost, but for most situations would not change the results very much, for two reasons. The first is that ownership is a much smaller percentage of total cost than the operating cost. The second reason is that the lower purchase price on a per space basis for shorter-lived vehicles tends to be offset by the need to purchase two or three of them over the life of a regular bus and hence the range of values tends to be narrow. Morlok, Bruun and Vanek (1) annualized the cost per space and found that it tends to be similar for all except the lowest cost vans and double-decker buses, both of which have substantially lower costs. If the preliminary analysis of a particular situation indicates the difference might be large, total cost can be used instead of operating cost: The relationship between total and operating cost for a regular bus, RB, is:

,

where TC is hourly total cost including ownership, OC is hourly operating cost, and CRF is the Capital Recovery Factor that can be found on many hand calculators or any finance or engineering economics textbook. The computation of Purchase Price for the shorter-lived vehicle must be converted into an equivalent value over life of the longer lived vehicle. For example, if the longer bus is to be used for 14 years and the alternative is to use minibuses that last only 7 years each, the equivalent value is computed by discounting the purchase price of the second vehicle P₂, at the end of 7 years, back to the present. This is then added to the first purchase, P₁:

,

where i is the discount rate used by the agency for capital investments. This is also the same rate used in the CRF computations. The equation for total cost of the minibus is then computed in the same fashion as the equation for the regular bus.

Breakeven demand

The breakeven demand is determined for two headways:

1. Short minimum headway: A minimum headway of 20 minutes is required on the route for level of service reasons. Costs therefore start at the minimum fleet and level of activity to have either three buses or three minibuses pass each stop in an hour, and increase after demand exceeds the minimum design capacity provided due to the minimum headway. As seen in Figure 1, breakeven occurs at approximately 50 passengers/hr for the high cost case, meaning that minibuses are cost-advantageous up to this level. The corresponding level is 100-125 passengers/hr for the low cost case.

{Figure 1 about here}

2. Long minimum headway: The minimum headway is 1 hour, with no additional service for peak periods. For this scenario, Figure 2 shows that breakeven occurs at approximately 15 passengers/hr in the high cost case, and 50-60 passengers/hour in the low cost case.

{Figure 2 about here}

This analysis shows the attractiveness of using minibus, especially if operating costs are the low end (i.e. 50% of RB).The number of minibuses per hour is 7 for the low-cost scenario and 4 for the high-cost scenario; in both cases, riders will perceive to be a higher quality of service relative to the base case frequency for the regular bus (i.e. every 20 and 60 minutes, respectively). Note that the above figures simplify the presentation of cost increasing with increasing demand in passengers per hour by smoothing the curve into a straight slope. In actuality, the cost curves for both types of buses are step functions, in which cost curves remain constant until the current capacity is reached and a new vehicle is added to the service; however, for the introductory purposes of this paper, the smooth curves were chosen.

Breakeven frequency

The assumption of no increase in ridership with increase in frequency is actually very conservative. A better level of service almost always leads to increases in ridership. Watts, Turner and White (8) found that the elasticity of demand with respect to increases in vehicle-kilometers (used as a proxy for increase in frequency), ranged between 0.2 and 0.4. The increase in ridership in turn generates more revenue and increases the operating budget. Therefore, when trying to approximately equate the operating cost of the regular bus and minibus services, it is reasonable to round up to one additional minibus instead of down.

In this analysis we take the RB frequency as fixed and estimate the frequency of MB service possible for approximately the same operating cost.For continuity, we repeat the short- and long-headway scenarios from the previous section, so that the baseline operating cost for use of RB is $277.74 per hour and $92.58 per hour, respectively.

The results are shown in Table 3. The frequency column shows the number of regular buses and minibuses that can be provided for a similar operating cost, while the cost ratio column shows the ratio of MB cost to RB cost for the given minibus operating cost and frequency of service.In the case of the short headway, a higher number of minibuses can typically be provided for only a small increase in operating cost.In the case of the long headway, the low starting point for frequency (i.e. RB on a headway of 60 minutes) leads to relatively high cost ratios for providing minibuses at a headway of either 30 or 20 minutes, depending on MB cost.However, this increase in cost may well be justified by the substantial improvement in service.For example, with MB cost at 65% of RB, the agency could double frequency from 1 to 2 services per hour with a 30% increase in operating cost.It is very possible that the revenue from the increase in ridership would fully offset the increasing costs in this case.

{Table 3 about here}

Area coverage analysis

A.Each route in the minibus network is the same length as the routes in the regular bus network, and the minibuses run at a shorter headway due to their higher average speed. This scenario prioritizes seat-trips provided over area coverage.

B. The routes in the minibus network, are 1.25 times longer than the regular bus routes, and both bus services run at the same headway. This scenario prioritizes area coverage over seat-trips provided.

Table 4 compares scenarios A & B with regular bus for various measures of performance.The results show that use of minibuses allows for a range of performance combinations, which in each case enhance the service compared to regular bus.Specifically, both scenarios provide 50% more routes than the RB base case, and in addition, scenario A provides 25% more runs per day, while scenario B provides 83% greater area coverage.

{Table 4 about here}

discussion

In some situations the minibus would improve level-of-service by higher operating speed and increased frequency with no increase in total operating cost. The primary theoretical reason not to make the conversion would be that inadequate capacity exists on some particular runs due to peaking characteristics. In practice, there are other institutional impediments, such as concerns over loss of higher paid jobs driving larger vehicles or fleet commonality.

In some situations the minibus would improve level-of-service by higher operating speed and an increase in frequency as well, but ultimately cost more to operate. Then there are several possibilities:

·The increase in passengers may be enough so that the operating ratio remains unchanged, but a net increase in subsidy is required to support more passengers.

·The increase in passengers is large enough that the entire increase is operating cost is paid for by passenger revenues.

·Improvements in service can often allow increase in fares, thereby either partially or fully covering any increases in expenses. For agencies looking for a means to increase fares without losses in ridership, minibuses deserve consideration.

The value of minibuses is particularly highlighted in the case of feeder routes to high capacity modes. They can either provide more frequency of connections to the trunk line, or extend routes farther into a community because of the higher operating speed, and thereby increase the area coverage of the network. Furthermore, feeder routes can double as community circulators, but circulators must provide frequent service in order to compete for short trips against the automobile. Minibuses make the high frequency required more affordable. Perhaps most important of all, the smaller size of the minibus might enable feeders to exist at all, since regular buses either can not operate on many residential streets or are not welcome.

A review of the American Public Transportation Association fleet composition statistics in recent years shows an increasing use of minibuses, so agencies are increasingly finding value in them. But the analysis here suggests that it is worthwhile for agencies to further review their existing routes using regular buses in order to identify additional candidates for conversion. As ITS technologies mature and become affordable to even small fleets, the value of minibuses can only increase. ITS will make more practical hybrid services that better address the needs of a portion of the disabled community. These riders will be mainstreamed with the rest of the ridership, at the same time saving the operating agency the expense of separate paratransit services.

conclusion

Acknowledgements

This article is based on work performed under grants from the Federal Transit Administration office of Technical Assistance and Safety, Delaware County (PA) Planning Department, and the Mid Atlantic Universities Transportation Center.The usual disclaimer applies.

References

1.Morlok Edward, Eric Bruun and Francis Vanek, 1997. The Advanced Minibus Concept: a new ITS-based Service for Low Density Markets, Report to the Federal Transit Administration, Grant No. PA-26-7000, University of Pennsylvania, Philadelphia.

2.Layton, Lindsey, 2000. “Springfield Circulator Buses Likely in for the Long Haul”, Washington Post,November 9, p.VA4

3.Wyman, Scott, 2000. “Cities Say Shuttle Gets Folks Moving,” Sun-Sentinel, June 4, p.4A-4B.

4.Staff, 1998. “SEPTA to Run Mini-Buses on Six Suburban Transit Routes,” Passenger Transport, August 10, V66

5.Staff, 1997. “Metro Dade is Growing Its Bus Service” Passenger Transport, May 5, V55.

6.Farwell, Randall G. and Eric Marx, 1996. “Planning, Implementation and Evaluation of OmniRide Demand Driven Transit Operations: Feeder and Flex Route Services,” Transportation Research Record 1557, pp.1-9, Transportation Research Board, Washington, DC.

7.US Department of Transportation, Federal Transit Administration, 1998. National Transit Database. Profiles of the 30 Largest Transit Agencies.

8.Watts, P.F., R.P. Turner and P.R. White, 1990. Urban Minibuses in Britain: Development, User Responses, Operations and Finances, Research Report 269, Transport and Road Research Laboratory, Crowthorne, Berkshire, England.

9.Vanek, F.“Breakeven Analysis for Comparison of Minibus vs. Regular Bus, Including Effect of Shifting Paratransit Demand to Minibus Routes.”May 5, 1995.http://www.virtualithaca.com/francis/MinibusReport.htm.Accessed July 30, 2001.

List of Tables and Figures

Table 1Comparative Characteristics of Minibus versus Conventional Bus Service

TABLE 2Assumed Characteristics of Large Bus and Minibus

Table 3Frequency of Minibus Service at Breakeven Cost Compared to Regular Bus Service

Table 4Comparison of Area Coverage for Regular Bus Versus Minibus

FIGURE 1Cost per Hour as a Function ofDemand for Regular Bus and Minibus Scenarios with Base Headway of 20 Minutes

FIGURE 2Cost per Hour as a Function ofDemand for Regular Bus and Minibus Scenarios with Base Headway of 60 Minutes

Table 1Comparative Characteristics of Minibus versus Conventional Bus Service

Minibuses, relative to standard buses, usually provide:

+higher frequency (or greater route density) and more demand-tailored routes

+higher operating speed, because of fewer stops for passsengers and greater manueverability

+lower environmental impact, due to lower noise level, less vibration, and smaller size

+greater service options and flexibility, e.g., stop on call, route deviation and service routes, greater driver attentiveness to needs of individual riders

+ability to provide service on streets and in neighborhoods where larger vehicles are excluded

+employment for more drivers although at a lower wage, and thereby creating a career ladder to large vehicle driving and management

+opportunity to contract out service

+ability to accommodate many ADA-entitled travelers, and others with special needs, reducing the need for costly paratransit service

+better servjce to these ADA-entitled travelers since shorter or no reservation time is required

+avoid negative publicity of large buses operating near empty

+can be used to test or build demand on new routes

-greater cost per space-mile

-shorter vehicle life (not always a disadvantage)

-greater congestion on crowded streets

-lower passenger volume capacity

TABLE 2 Assumed Characteristics of Large Bus and Minibus

Regular Bus (RB)

Hourly Operating Cost per Revenue Hour$92.58

Hourly Operating Cost per Revenue Mile$9.08

Revenue Operating Speed16.3 k/h (10.2 mph)

Spaces50

Design Load Factor0.8

Minibus (MB)

Hourly Operating Cost per Revenue Hour(0.5-0.8) x $92.58

Hourly Operating Cost per Revenue Mile(0.5-0.8) x $9.08

Revenue Operating Speed1.25 x 16.3 k/h (10.2 mph)

Spaces20

Design Load Factor0.8

Table 3Frequency of Minibus Service at Breakeven Cost Compared to Regular Bus Service

 

	Fraction of RB Cost	RB Frequency	MB Frequency	Cost ratio
Short headway	50%	3	7	1.17
	65%	3	5	1.08
	80%	3	4	1.07
Long Headway	50%	1	3	1.50
	65%	1	2	1.30
	80%	1	2	1.60

Table 4 Comparison of Area Coverage for Regular Bus Versus Minibus

Figure 1 Cost per Hour as a Function of Demand for Regular Bus and Minibus Scenarios with Base Headway of 20 Minutes

Figure 2Cost per Hour as a Function ofDemand for Regular Bus and Minibus Scenarios with Base Headway of 60 Minutes