Paratransit Guide Supplement
Planning and Performance Monitoring
by Richard Schultze
NOTE: This supplemental section provides more in-depth treatment of the section on "Planning and Performance Monitoring" found in pages 26-30 of "Paratransit for mobility-impaired persons in developing regions: Starting up and scaling up," a guide prepared by Access Exchange International. The opinions expressed are by the author and readers will want to adapt this material to their local conditions.
This section goes beyond the brief introduction to planning and performance monitoring in the paratransit guide posted on our website. It describes additional criteria, and factors to take into consideration when planning cost-effective and appropriate demand responsive transportation (DRT) which is accessible to seniors and passengers with disabilities.
Defining demand responsive transportation
DRT is provided under a wide variety of names, including
- dial-a-ride service
- disabled services
- door-to-door service
- curb-to-curb service
- shuttle service
- service routes
Provided by different types of vehicles, including
- accessible vans
- shared ride taxis
- non-motorized vehicles
- and many others
The primary characteristic that distinguishes demand responsive transportation (DRT) from conventional fixed route transit is that service changes daily in response to daily changes in demand for the service.
DRT can take the form of several major service models:
- fixed route deviation approach: relatively small changes are made to established fixed routes (called "route deviation" or "flex-routes") to accommodate individual trip requests on a limited basis, for example up to one kilometer
- point deviation approach: specifies time points but allows for changes in schedule between these time points
- "taxi" approach: there are many and major changes every day to accommodate many trip requests
The fixed route deviation approach uses a conventional fixed route service as the base, with specified time points (where and when the vehicle needs to be) and specified routes (or maybe flag stops), but allows the vehicle to deviate off-route (up to 1 kilometer, for example), to accommodate people who cannot get to the route. With advanced requests from riders, the bus goes off route, picks up or drops off the rider, comes back to the route at the exit location, and then resumes the scheduled service.
No one can anticipate the demand for a deviation, and when and where it will occur. Thus the schedule for this type of route has to include enough time at the end of the route (called "schedule recovery time") so that the vehicle can get back on schedule. When a deviation does occur, it is likely that the bus will be running late for awhile, until it can recover at the end of the line.
The point deviation approach uses a set of specified time points (where and when the vehicle needs to be) but allows for multiple deviations (e.g., up to 1 km) between the time points. These deviations require a prior request. There is increased flexibility over the fixed route deviation approach, but it is harder to operate and to schedule.
The "taxi" approach does not mean that the service is taxi service, but does mean that the service is highly variable from day to day, with many individual trips, and perhaps some group trips, as well as picking up and dropping off riders along the way. The service cannot be shown on a schedule brochure, since there is no semblance of a fixed route or fixed time points. This is the most flexible service there is, but it is the most difficult to schedule and administer. Some consumers may find it difficult to understand, but since it is like a taxi, most will understand how it to request and plan for a trip.
Beyond these major options, there are many variations on this theme, with DRT able to have any or all of the following main characteristics, depending on the local situation.
Demand-response transport operators may
- provide vehicles with wheelchair securement areas and tie-downs, wheelchair lifts or ramps for electric and manual wheelchairs, or they may opt only to serve passengers who do not need to remain in their wheelchairs during travel
- provide public transit DRT (open to the general public) or social services DRT (open only to the clients of the particular social services agency)
- serve only disabled and elderly riders without mixing with the general public, or instead combine riders as needed
- provide curb-to-curb, door-to-door, or through-door service, or some combination depending on the rider’s needs
- require advance reservations, allow same-day requests, or both
- provide direct service from home to destination for single trips, or more circuitous routing to try and carry more than one person at a time, or attempt to group trips, or attempt to provide time transfers at nodes to allow for longer distance trips
- provide transport only within zones or instead cover a larger area
- focus on transport to certain locations or to many locations
- provide feeder service for transfers at key stops and stations on conventional fixed route services
- provide attendants for elderly or disabled passengers
- use smaller vehicles than conventional fixed route service
- provide a lower fare (to reflect the income level of many of the disabled and elderly, and rural population) or a higher fare to reflect the higher costs of DRT
- allow any trip purpose or limit travel to one or a few trip purposes
- operate every day or just weekdays or once a week
- operate all day long or limit the hours of service
In addition there are many smaller decisions that need to be made. These include
- establishing a process for a driver to leave the bus and help a disabled or elderly rider to walk or role to the vehicle and then get on or off the vehicle, as well as securing the rider in the seat, or tying down the wheelchair while securing the rider
- determining what types of wheelchairs will be allowed on board
- establishing criteria for drivers to use when determining where to stop to allow boardings and alightings, especially where the road and sidewalk (if any) are dangerous
- defining "no-shows" and trip cancellations
- how long to wait for a person before leaving and declaring them a no-show
- defining the “window” of time within which a rider needs to be ready to be picked up
- dealing with riders that are late coming out from medical appointments
- establishing a process to appeal temporary or permanent suspension of the right to use the service
- allowing caregivers to accompany elderly or disabled passengers
- allowing service animals such as dog guides for the blind
- equipping vehicles with other amenities for the disabled and elderly
- providing bike racks so people can extend their range of travel beyond the service area
- incorporating package delivery
- carrying students to schools
- providing service changes to accommodate for rural market days (and the need to transport goods), or to provide for seasonal, monthly and weekend movements of workers to and from the countryside in order to work in the cities
- providing service changes to accommodate religious holidays and pilgrimages
- providing service changes to accommodate seasonal holiday travel
- allowing live animals, goods, and personal luggage and packages on the vehicles
- establish policies concerning allowing riders to stay on board the vehicle when the driver has to leave to take a personal break
- establishing procedures to avoid leaving a person without a ride back
- having information about each special needs rider in the dispatcher's office for easy access during emergencies
- stopping service at jurisdictional boundaries or going further to accommodate the demand
- dealing with and reporting incidents on the vehicle
- establishing policies concerning the mixing of sexes and the observance of cultural norms on the vehicle
- securing medications when transporting a disabled or elderly person who is carrying medications or who needs medications at the destination
- providing travel training for disabled, elderly and illiterate riders
“Family of transportation services”
There are many elements to the “Family of Transportation Services”. The DRT provider should occasionally review this list of potential service types to see if perhaps one or more of them, or some variation, may be appropriate for implementation.
The many different elements may include:
- fixed route transit
- feeder service
- flex-route options
- demand responsive service
- complementary accessible paratransit
- coordination with specialized human services agency to provide paratransit service that could be curb-to-curb, door-to-door, or door through door
- special shuttle service
- volunteer driver programs (possibly reimbursing volunteers for their costs)
- use of transit pass or voucher programs
- gas-voucher programs
- travel training
- carpool and vanpool programs
- vehicle-sharing programs
- use of accessible taxis
- education, information and outreach to passengers
- private-vehicle loan, purchase, or donation programs
- linkage between pedestrian and bicycle routes
- simplified one-stop access where passengers may make trip requests
This “family of transportation services” uses the empowering capabilities of information technology. Such technology include GPS/AVL, mobile data terminals or tablets, on-board security cameras, scheduling software, electronic fare boxes, smart fare cards, automatic passenger counters, radio or cell phone communication with a dispatcher, electronic destination signage, and GIS and traveler information systems at stops, via the internet, or via phones.
At one time public transportation was just "fixed route" (express, local, crosstown, circulators, and paratransit) or "demand responsive" services (usually for smaller urban and rural areas), with little reliance on information technologies. Transit systems are now learning to apply new approaches to increase mobility options, fill service gaps, make public transit more relevant, decrease their operating costs, and partner with other agencies.
While there is good information describing successful approaches to incorporate the various elements of “the family of transportation services”, and ways to measure those services, there is not much information about standards for those measures. Transit systems will have to review the literature, contact other systems that have implemented the services, and then track those new services over time.
DRT systems can benefit by initiating and participating in coordination and mobility management efforts. This is when public transit systems, social services agencies, and private operators work together to identify gaps and overlaps in service, needs for funding and vehicles, and needs for more technical assistance. For example, such assistance could include contracting with one provider to carry another provider’s trips when needed. This kind of cooperation, even in a competitive arena, can help providers stretch their funding, carry more riders, and be better at what they do.
Once the defining characteristics of the local DRT system are determined, management should develop clear policies and procedures that incorporate those decisions.
Forecasting ridership and revenue
Forecasting ridership and revenue is an iterative process, and in many ways is just good common sense and examining your assumptions. Initial assumptions are made about service area and times, total demand vs. ridership to be carried, fares generated, vehicles needed, service costs, etc. This process may need to be refined until management feels that all of the assumptions make sense, are realistic and are consistent.
The most important data to have is population, hopefully broken down by disabled persons and elderly persons, and by residents and workers in various geographic areas. If the national or local governments have such data, then review it for reasonableness and use it. If no such data exists, then it will be necessary to gather together professionals from the community that may have some knowledge of population and demographics, and then develop a set of best planning estimates. As a general rule, it is best to focus service on where the largest concentrations are of disabled and elderly population, and on the areas with the largest number of residents and workers. In those countries where household income and car ownership data are known, it is best to concentrate on the poor neighborhoods and the areas with low auto ownership.
Given the data on population, then the easiest approach to forecasting ridership is the comparable service method. This consists of comparing a proposed service, service area and population with an existing similar service, service area and population. This can be done in the same area or data can come from a similar urban or rural area elsewhere. However, this cannot be done if there are no comparable data sets. Ridership per capita can be used as a reasonable estimate.
The planning staff of some cities and regions may also have access to trip origin and trip destination data. This is useful because it represents flows of people going to work, and for other trip purposes.
Depending on its budget, management may be able to conduct surveys of neighborhoods to obtain information about travel patterns and willingness to use transit. However, such surveys can be expensive.
There are computerized travel demand forecasting models that are available commercially, but their use is more appropriate for larger fixed route bus systems and for rail systems.
In general, forecasting ridership for DRT is quite difficult and not very predictive. The best approach is to make the best estimate, implement service, collect data, and make service corrections.
Measuring DRT service
Both large and small transport systems need to collect and report basic transportation data. The following data is considered essential, in order that the transportation system owner or manager will know what is going on with the transportation program. The following paragraphs describe several key indicators that you should use to measure your system.
Indicator #1: One-way passenger-trips (OWT)
How much annual ridership should you be carrying? This is a function of the service type, amount of service, service area, demographics, land use patterns and intensities, quality of service, and the markets you are serving.
Is ridership flat, increasing, or decreasing? What are the reasons for this? Should there be changes in you service or fares? Are there changes in the demographics or economics in the service area? Are the reasons for these changes understood? Is these trends acceptable, or is action necessary to address them?
Ridership should periodically be evaluated by category and compared to the service area population. Riders could be analyzed in terms of the available of personal automobiles, income level, age, gender, employment status, minority, education level, disability, trip purposes, or fare paid.
OWT is one person making a one-way trip. This indicator shows the volume of trips carried by the system.
For example, Ms. Smith going from her home to downtown is a one-way passenger-trip and coming home is another one-way passenger-trip for a total of 2 one-way passenger-trips. Ms. Smith going from home to the downtown then to a doctor and then home is 3 one-way passenger-trips.
Unlinked passenger trips are counted each time a passenger boards a vehicle no matter how many vehicles they use to travel from their origin to their destination.
Linked passenger trips do not count transfers and are the full trip from origin to destination on the transit system. Even if a person must make several transfers during a journey, the trip is counted as one linked trip on the system.
In other words, an individual or rider is not the same as a trip.
The OWT data may come from:
- a driver’s marked-up manifest
- dispatcher or scheduler records
- GPS or MDTs or tablets
- point checks
- maximum load counts
- ride checks
- boarding counts
- farebox readings
- onboard surveys
- point checks
- vehicle utilization charts
- passenger surveys
The total OWT data should be recorded and then shown on a series of tables and graphs for each day the service operates, then added up by month, quarter and year, as well as by day of the week.
The data gives you an idea of how the transportation service is used during a particular period. You want this number to be increasing over the year, and from year to year. If the number starts to decrease, then there is a problem which must be understood and addressed. The data will help in scheduling, budgeting, cost control, grant preparation and reporting, public relations and marketing, and an overall appreciation of the transportation activity of the system. The data, tables and graphs should be made easily available on a regular basis to the transportation and financial staffs.
Indicator #2: Individuals served
This indicator counts unduplicated people transported and shows how many lives are impacted by the transportation.
For example, Ms. Smith is 1 individual served, no matter how many trips she takes during a day, month or year.
The data comes from the driver’s marked up manifest or the dispatcher and scheduler’s records. You want this number to be increasing over the year, and from year to year. If the number starts to decrease, then there is a problem which must be understood and addressed. The data will be useful in public relations and marketing, and in showing a positive community impact.
Indicator #3: One-way passenger-trips by funding category and by trip purpose.
This data is important for invoicing and grant reporting.
For example, if Ms. Smith travels from home to downtown for lunch this is one trip, and if she then goes to the doctor and home there are 2 “medical” trips.
This data comes from the dispatcher/scheduler’s data. The data by funding source and trip purpose will add up to the total OWT data described above. Similar to the regular OWT data, the purpose and funding source data should be recorded and then shown on a series of tables and graphs. The data will help in ensuring that you are meeting the obligations of funding sources, and in public relations and marketing.
Indicator #4: Vehicle-kilometers of service by funding category
One vehicle going one kilometer is a standard unit of measure when determining the cost of service and the need for vehicle maintenance.
For example, vehicle A will leave the base of operations and may travel 100 kilometers during the day transporting people. That is considered 100 vehicle-kilometers. The number of vehicle-kilometers is taken from the odometer by subtracting the beginning odometer reading (taken before leaving the base/vehicle storage area for the first trip of the day and written by the driver on the manifest) from the final odometer reading (after arriving at the base/vehicle storage area after the last trip of the day and written by the driver on the manifest). This will include the kilometers going from the base to the first pickup, going between a drop off and a pick up, and going from the last drop off to the base. Other distance travelled, such as going for a package pickup, going to a maintenance shop, etc, are not included in the vehicle-kilometers.
The total vehicle-kilometers for each vehicle should be recorded and then shown on a series of tables and graphs for each day the service operates, then added up for the whole fleet and shown by month, quarter and year, as well as by day of the week.
The data is important in invoicing funding sources for costs, determining the costs of services in general, budgeting, and vehicle maintenance. You want this number to be close to the forecasted vehicle-kilometers used in the budget. If it goes over, there will be budget problems. The data will help in scheduling, budgeting, cost control, grants preparation and reporting.
Indicator #5: Vehicle-hours of service by funding category
One vehicle going one hour is an important unit for costing service and for determining the required staffing of your system.
Similar to vehicle-kilometers, for example, vehicle A will leave the base and may travel for 6 hours during the day transporting people. That is considered 6 vehicle-hours. The number of vehicle-hours is the time between the beginning time (taken before leaving the base/vehicle storage area for the first trip of the day and written by the driver on the manifest) and the final time (after arriving at the base/vehicle storage area after the last trip of the day and written by the driver on the manifest). This will include time going from the base to the first pickup, going between a drop off and a pick up, and going from the last drop off to the base. Other time mileage generated, such as driver break time, going for package pickup, going to a maintenance shop, etc, are not included in the vehicle-hours.
The total vehicle-hours for each vehicle should be recorded and then shown on a series of tables and graphs for each day the service operates, then added up for the whole fleet and shown by month, quarter and year, as well as by day of the week.
The data is important in invoicing funding sources for costs, determining the costs of services in general, budgeting, and vehicle maintenance. You want this number to be close to the forecast vehicle-hours used in the budget. If it goes over, there will be budget problems. The data will help in scheduling, budgeting, cost control, grants preparation and reporting.
Indicator #6: Productivity ratios
There are three groups of productivity ratios: efficiency, effectiveness and cost-effectiveness.
These ratios are useful management and service costing and planning tools.
These ratios are easy to calculate, and use the total number of one-way trips, vehicle-kilometers, vehicle-hours, and operating costs for any particular period of interest (month, quarter, annual).
You should track these ratios over time and compare your ratios with those of your peer transportation companies or agencies to see if your numbers are reasonable.
Efficiency indicators show the amount of resources needed to produce the operations. Typical efficiency indicators are:
- Costs per vehicle-mile
- Costs per vehicle-hour
Cost/mile and cost/hour are very useful and informative. A transit system can compare these measures over time to: (1) see trends, (2) compare with that of peer transit systems, (3) to evaluate service cuts, additions, realignments, or new types of service, and (4) to compare with other competing modes and systems in the area.
It is important to use both measures at the same time in order to have a full picture. Each route or service may have different average speeds (kilometers/hour). When using only one of the measures, you will likely see different rankings or evaluations with one measure compared with the other measure.
Example: compare two routes A and B:
Average speeds: A-20 kph (more urban), B-30 kph (more rural)
Daily vehicle-hours: A-12, B-12
Daily vehicle-kilometers: A- 240, B-360
Daily costs: A-$600, B-$720
Which is more cost efficient?
A: Costs/Kilometers =$600/240=$2.50, Costs/Hours = $600/12=$50
B: Costs/Kilometers= $720/360=$2.00, Costs/Hours =$720/12=$60
The two ratios of operating cost/vehicle-kilometers and operating cost/vehicle-hour help you understand how the costs of actually providing the service are changing over time. You want these ratios not to increase much from year to year to show that the service is not becoming too costly. If the ratios increase too quickly, then there is an issue about the actual cost structure in your transit system, and this would need to be addressed quickly. If you can get these ratios to increase, due to better scheduling, wage negotiations, and achieving economies of scale due to service increases, then this is good.
Effectiveness indicators show rides generated by the operations. Typical effectiveness indicators are:
- Passenger trips per vehicle-mile
- Passenger trips per vehicle-hour
Trips/mile and trips/hour do not mention or depend on costs. Use both measures at the same time.
The two ratios of OWT/vehicle-kilometer and OWT/vehicle-hour help you understand how many trips you can carry on the service that you provide, and show you how effective are the daily schedules that you develop. You want these ratios to increase from year to year. The data can be used to rank existing and proposed services, to help guide you in adding or cutting services.
Cost-effectiveness indicators show how much it costs to generate the rides. Typical cost-effectiveness indicators are:
- Costs per passenger-trip
The ratio of operating cost/OWT helps you understand basically how much it costs to transport each trip. You want this ratio not to increase much from year to year to show that the service is not becoming too costly or that you are carrying fewer trips for your costs. If the ratio increases too quickly, then there is an issue about the actual cost structure at your transit system, or an issue about the number of trips you can carry, or both, and these would need to be addressed quickly.
Cost effectiveness (Cost/Trips), cost efficiency (C/Hours, C/Kilometers), and service effectiveness (Trips/Hour, Trips/Kilometer) are mathematically related:
Cost/trip = (cost/hour)/(trips/hour) and Cost/trip = (cost/mile)/(trips/mile)
It is good practice to use all three types of productivity measures when evaluating service, doing trend analyses and doing peer comparisons. Each indicator gives one view of the service, but not a complete view.
Indicator #7: Quality of service data
Quality of service indicators show the service from the point of view of the riders. They show what riders care about and what will impact their decisions to use your service. A discussion of the indicators is found following the list below.
Typical quality of service indicators include:
- On-time performance (percentage of trips between 0 and 5 minutes late)
- Safety (preventable and non-preventable accidents per 100,000 vehicle-kilometers)
- Breakdowns (number per month)
- Wheelchair lift working (percentage not working per month)
- Travel time (transit travel time as a percentage of auto travel time; maximum time on board vehicle)
- Convenience (percentage of population within a kilometer, or fraction of a kilometer, of stops, time between vehicles, hours of operation, percentage of trips requiring transfers)
- Equity (percentage of transportation disadvantaged population with transit vs. non-transportation disadvantaged population)
- Trip denials (number per month)
- Wait lists (number on the list)
- Quick response (time between when a rider calls in for a trip and when they get picked up; particularly for same day service)
- Accessibility of service (percentage of vehicles that are wheelchair accessible; number of wheelchair trips carried)
- Comfort (percentage of vehicles with working air-conditioning or heating, days between vehicle cleaning)
- Pricing (fare as percentage of cost; fares compared to other modes)
- Customer satisfaction (percentage of passengers that rate the system good or excellent; number of complaints)
- No-shows and cancellations
- Maximum load factor (ratio of all people on board divided by the number of seats, at the point of maximum crowding)
- Mode split (transit trips as a percentage of all trips taken in the region)
This data gives a good view of how the riders and community perceive your service.
The quality of service measures can be calculated for any particular period of interest (month, quarter, annual), and should be presented monthly in order to detect trends at their beginnings.
You should track these measures over time and compare yours with that of similar or equivalent agencies providing transportation, to see if your numbers are reasonable.
The accidents per 10,000 vehicle-kilometers, malfunctioning air conditioning or heating per 10,000 vehicle-kilometers, and malfunctioning accessibility equipment (lift, tie downs, etc.) per 10,000 vehicle-kilometers helps you understand if the service is safe, well-maintained and comfortable. Accidents are considered to be those that are preventable, and outside the responsibility of the driver. You want these ratios to go down from year to year.
The measure of the average amount of time a rider is on the vehicle helps you determine if the service is as direct as it can be and well scheduled, so that riders do not have to spend an inordinate amount of time on board. This is important especially for the elderly and disabled. You want this ratio to go down from year to year, although there is a conflict with the ratios of one way trips per vehicles-mile and one way trips per vehicle-hour, since improved trips per hour may make the routes more circuitous to get more people on board.
The measure of total trip requests that are denied by the system and cannot be carried (after trying to work with the person to see if the trip times can be changed to fit in with the schedule) helps you understand better about the demand that is not being met and the frustration level of those who cannot get a ride. This can be an indication of the need for more efficient scheduling to carry more trips with the resources available. Yet it may also be an indication of a growing need for transportation, as the population ages, as people are still out of work, as people have more medical needs that require visits to the doctor, or similar reasons. These denials of trip requests very much tend to underestimate the true demand, because, after a short time, potential callers just do not try anymore since they know there is a good chance that their trips cannot be carried. While we all want this number to go down from year to year, it usually will not do so, especially if you are running a service that people really want and need to use.
The measures of no-shows and cancellations by riders need to be monitored on a system basis and on an individual basis. Riders no-show for many reasons, including being sick (and not being able to call in or not calling in soon enough), forgetting about the trip, being confused about the trip time, booking several rides at once in the hope of one coming through, or just being irresponsible. Cancellations occur when a rider senses that he/she will not be able to take the ride, for one of the above reasons, but is able to call the system soon enough. Both no-shows and cancellations detract from an efficient schedule and basically take viable seats from others who would be ready for the ride. You want these numbers to go down from year to year. Strong policies about no-shows and cancellations are warranted.
The measure of percent of trips that are late (at the destination) helps you determine if the service is reliable, which is very important to the rider. The reasons may be incorrect information given to the scheduler, inefficient scheduling, or breakdowns with the equipment. You want this ratio to go down from year to year.
The measure of complaints per 10,000 vehicle-kilometers helps you understand if the riders and the general public (that may be viewing the vehicles and interacting with them in traffic) believe that the service is of good quality and safe. You want this indicator to go down from year to year. You should also report out any complaint, suggestion and compliment to the upper management of the system so they have a running evaluation of how the community and riders view the system. Note that compliments are an important indicator, not just complaints! Drivers and other staff should be commended when they receive compliments from passengers.
Management needs to have a comprehensive set of service productivity, service quality and business indicators and standards against which to measure and monitor the service. Periodically (at least quarterly), management should collect the data stated in the indicators and compare them with the standards. This will allow management to evaluate and rank the existing services, to show which services should be retained, improved, or discontinued. It may be hard to collect accurate data, especially at first. Obtaining data from rural services may be especially difficult. However, experience has shown that management needs this kind of data to operate cost-effectively and to retain riders and revenues.
Over time, a system can generate data on how the indicators change over time. That can help management establish performance standards which are benchmarks against which to measure the service.
It is also beneficial for a system to compare itself to other transit systems and businesses of comparable size, operations and passenger profiles. This will allow for a peer review and can also facilitate setting absolute standards. Of course the ease of comparing your service with others will vary from one country to another and may be impossible if you starting the first paratransit service for passengers with disabilities in your city.
Management should use these indicators to set quarterly and annual systemwide goals to help push the system to continuously improve. In addition, management should report several basic indicators (passengers/vehicle-kilometer, passengers/vehicle-hour, cost per passenger, etc.) for each service, in order to differentiate the strong from the weak services and focus service changes and improvements.
DRT services tend to have lower riders/mile and riders/hour, and higher costs/rider, than do conventional fixed route services. This is to be expected, since the disabled and elderly tend to be spread throughout the area, and each of them take significantly more time to load and unload. This is also to be expected since the rural population is spread over wide areas and it takes a disproportionate amount of time to serve them. Nevertheless, the productivity numbers for DRT are significantly better than they would be if the services were to be provided by conventional fixed route services.
Once the decisions are made about the type of service to operate, those decisions need to be formalized into a set of service design criteria. The service design criteria will serve as a reference for the service planning, scheduling, dispatching and management and will clearly state what kind of service is desired. Periodically, management needs to review those service design criteria to determine if they need to be updated. And, periodically, management needs to compare the service on the street with the criteria, to identify unmet needs and to determine if the criteria are being followed.
Ongoing service evaluation process
Proposed service indicators (and any specific standards or goals that you may set for them) provide a tool for the planning and operating of transit service. These indicators can be used to evaluate performance once the routes comprising this system are in operation.
In some cases, design and performance standards may not be met because of cost, personnel, equipment, or other constraints. Every effort should be made to provide service that is in conformance with these standards. The process of identifying routes or portions of routes that either exceed or do not reach these standards should be used to guide decisions in response to decreases or increases in funding for service.
The transit system should regularly evaluate the performance of individual services based upon the range of service measures, especially those dealing with productivity and quality of service.
Those services performing marginally or unsatisfactorily will be scrutinized for ways to improve productivity. Services that are “good” may be considered for improved service.
Data determining the total number of passengers, the number of passengers per vehicle mile and vehicle hour, and the revenue-to-cost ratio, for each route and service category, should be collected on a monthly basis. Other performance indicators such as schedule adherence should also be monitored. Performance information for the system and for each individual route should be produced on a monthly basis. This report should show the actual productivity measures for that particular month and be the basis of an on-going evaluation of individual vehicle routes or service areas.
Route evaluation, if your service is based on routes, should follow a two-step process. The first step is to identify how each service performs in terms of ridership, cost effectiveness, efficiency and effectiveness, and quality indicators. Once this screening process is completed, the second step is a detailed evaluation of each service that is either not performing up to standard or is performing well above average. These routes could be subject to a number of actions including changing the frequency or times of service, consolidating or eliminating routes, or modifying the way you publicize your services.
The periods of time for which performance indicators are calculated will relate to internal and external reporting requirements. Typically, they should be compiled on a monthly basis. Comparisons with the previous month, the same month for the previous year, and with a two or three year running average should be included to identify trends.
Unacceptable routes should receive a more detailed analysis in step two. At this level in the performance evaluation process, a route’s performance in all standards is analyzed. This detailed evaluation should examine performance by time period, day of week, and route segment. A variety of measures should be considered for action including redesign, schedule reduction, schedule increase, marketing activity, and other actions.
Each route should be reviewed annually. Routes should be reviewed more often if ridership trends are negative, special requests for services are received, or other special circumstances are noted. The evaluation process should also include routes with good performance. These should be candidates for more frequent service or some other kind of service increase.
Newly established transit routes at specified levels after an initial period during which they ramp up service. Service changes should be made no more frequently than every several months, so as not to confuse the riders and operations personnel.
DRT services tend to have low riders/mile and riders/hour, and high costs/rider. This is to be expected since the disabled and elderly tend to be spread throughout the area, and each of them take significantly more time to load and unload. This is also to be expected since the rural population is spread over wide areas and it takes a disproportionate amount of time to serve them.
Obtaining feedback about service
Management needs to actively seek public input on the type of service it offers, the span (days and hours) of service, where the service goes, the fares, the types of vehicles, how well it serves the disabled and elderly, etc. Unless the transit system works hard to meet the real mobility needs of the rider, experience has shown that transit riders will choose not to use other transport modes when possible. Input should not be solicited just from riders, but also from business and government agencies whose workers use the service, medical facilities whose clients and workers use the service, and the non-riding public.
In many places, it may be difficult to obtain input from disabled and elderly riders, especially those that are poor, illiterate and disenfranchised, without access to convenient and widespread media. Nevertheless, this input is vital, and strong efforts need to be made to obtain it.
There are several methods that can yield public input:
- public input meetings in neighborhoods or public areas
- posting the phone number, mailing address, email address and website of the transportation system, and encouraging public comments
- advertising in local newspapers and magazines, trade journals, union newspapers, newsletters of places of worship, movie theaters, TV, radio, and on the outside and inside of the vehicles
- random surveys (household, telephone, mailed) and focus groups
Management should actively solicit driver and dispatcher feed back on service and operations, since those persons are in daily contact with the service and are professionals.
Management also needs to monitor how the service is actually operated on a day-to-day basis (as distinct from the service design). This is essential to good service planning, since planning and implementation can often yield two different products.
Management needs to provide clear, comprehensive and written policies and procedures that specify driver behavior. Drivers that violate these requirements need to be quickly held accountable through verbal and written documentation, verbal correction and retraining, disciplining and even termination. Similarly, drivers that uphold these requirements should be rewarded and commended.
Drivers that are monitored while they work tend to do perform better than drivers that are not monitored. When drivers do their job well, the service can be more safe, pleasant, punctual and respectful of the riders. When drivers do not do their job well, there can be speeding, reckless driving, not completing the route or not picking up all the passengers, falsification of records, skimming fares, not requiring some people to pay fares, being rude, threatening or prejudicial to riders, conducting other business from the vehicle, driving under the influence of drugs or alcohol, talking bad about the system, damaging the vehicle, etc.
There are five major ways to monitor drivers: (a) complaints and compliments from the riders, (b) complaints and compliments from non-users, (c) “secret riders” employed by the transportation system, (d) street supervisors employed by the transportation system, and (e) on-board cameras.
Riders of the service can be the most important source of feedback about the drivers. Unfortunately, in some cases, management may not value the observations and opinions of riders, and may not take them seriously, since many riders are not literate or educated, and many are poor or old or disabled, and without powerful connections. Yet wise management can benefit from rider observations, since the riders do spend a lot of time riding the vehicles and are often keen observers of the driver and service quality.
In order to receive the rider observations, management has to put in place a range of methods for the riders to send feedback to management. Such methods include: posting phone numbers to call, email addresses to contact, websites that have a means to provide a comment, having a rack with comment cards that can be mailed in to the main office, allowing for walk-ins to express their observations at the main office. It is important that the drivers cannot trace the comments back to the individual riders, because fear of retribution can stop comments from being given. If possible, persons that submit comments should be contacted later to let them know how the comments were handled. This may be difficult since many people may not want to identify themselves. But, it is important to note that just because a person wants to remain anonymous, their comment is still valid and worth reviewing.
Non-users of the system, including pedestrians, and drivers of personal autos, trucks and other vehicles for hire, do observe the transit vehicles in operation and see how the transit vehicles interact with other vehicles and pedestrians. The outside of each transit vehicle should clearly display the bus number ID, a telephone number and a website for people to contact.
The transportation system can employ “secret riders” to ride the vehicle incognito, and report back to management. Secret riders need to be trained by management on what to observe and how to record the observations. This can be very effective, as long as the identity of the secret rider is not compromised.
Larger transportation systems may employ street supervisors to monitor the service. These professionals should ride the vehicle (even though the presence of the street supervisor will make the drivers perform better), follow the vehicles in a chaser car, observe the loadings and unloadings at stops and terminals, etc. These trained professionals will have a long checklist of actions to observe.
Some transportation systems may be able to afford on-board cameras, to record the driving and behavior of the driver, as well as to monitor the behavior of the riders and to show any criminal activity by the driver, rider or non-rider. Cameras, the associated computers, storage and on-going equipment maintenance can be quite expensive, but may be worth it in some cases.
Overall, feedback and reviews of service should be viewed as a necessary and valuable tool to continuous improvement, even though sometimes painful and threatening.
Date of this supplemental material: June, 2013
All of these references can be downloaded from the Internet for free, unless so noted.
- Assessing the Need for New Transportation Services in Rural Communities, Easter Seals/Project Action, no date.
- Family of Transportation Services, United We Ride, Federal Transit Administration web site, 2007
- Innovative Practices in Paratransit Services, Easter Seals/Project Action, 2002; sections on scheduling, dispatching, reservations, service quality (telephone performance, travel time, on-time performance, trip denials); www.projectaction.org, then click on “resources and training”, then click on “Innovative Practices in Paratransit Services” to download. Or, more simply, just Google Innovative Practices in Paratransit Services and download.
More complex references:
- TCRP Report 124: Guidebook for Measuring, Assessing and Improving Performance of Demand Response Transportation, 2008.
- TCRP Report 136: Guidebook for Rural Demand Responsive Transportation: Measuring, Assessing and Improving Performance; 2009
- Public Transport in Developing Countries; Richard Iles; Elsevier; 2005; [can be purchased over the Internet]
The TCRP documents can be found at: www.trb.org, then publications, transit (TCRP). TCRP stands for Transit Cooperative Research Project.
(This Paratransit Guide Supplement added in November, 2013)