[IMAGE]MET7p36Full.jpg[/IMAGE] 

Transit cost efficiency, synonymous with productivity of vehicle and crew resources, is a growing concern for many agencies of the public and private sectors. In spite of the emergence and growth of a strong social conscience and, hence, political goal to be more earth-friendly to protect our natural resources, North-American transit agencies have had to operate on ever-shrinking operating budgets.

The years of federal operating subsidies are long gone, and state/local operating subsidies are at risk now more than ever. The challenge is clear: public transit agencies must reduce their operating budgets; the means are simple: operate more efficiently, from an operating cost perspective, or undertake drastic service reductions.

Cost efficiency must not, however, occur at the expense of service effectiveness. System connectivity, system punctuality and passenger loads are only a few of the service effectiveness elements to which patrons are most sensitive to.

Service analysis

In broad terms, a Service Efficiency Analysis (SEA) consists of a systematic review of how transit services are planned and scheduled. The dominant task in this process is a review of existing labor rules, scheduling practices and efficiencies, and a review of production schedules. This review focuses on existing scheduling approaches and efficiencies, with an objective of identifying areas in which the agency may realize efficiency gains through updated practices.

The revised practices proposed typically result from joining local operational considerations with the diversity of international innovations and scheduling practices of varying locations, such as Europe, North America and Australia.

It is not uncommon for an SEA to include a peer system study and/or a labor negotiations component. The former benchmarks the agency in terms of industry standards, while the latter highlights labor rules most constraining to efficiency gains, which should be part of the next labor negotiations.

Of course, the participation of senior members of the planning, scheduling and operations teams of the agency is crucial; such staff holds knowledge critical to a meaningful, successful and implementable SEA.

 [PAGEBREAK] 

Transit service deployment

The most important concept of transit efficiency is the binding relation between the processes of transit service deployment. The service plan must give consideration to scheduling constraints and objectives, which in turn must address daily operations requirements. In other words, each department must diligently think down the line or have an extended vision.

Unfortunately, extended vision is progressively lost as recent years have seen the gradual segregation of the processes inherent to transit service deployment. Even within each process, further segregation is not infrequent. Timetable development, vehicle blocking and crew run cutting may involve three different staff.

The segregation of processes results in disconnection and narrows down the vision of efficiency and effectiveness. Global efficiency and effectiveness are lost, replaced by a series of local exercises with limited consideration to the down-line impacts. Regrettably the sum of multiple local optima is never better than the global optimum.

The role of the SEA is to document the benefits of a more integrated approach and assist in re-establishing an integrated method of planning and scheduling. Applications of this include cross-training of staff and grouping staff by operating division rather than by process.

A review of Standard Operating Procedures (SOPs) is amongst the building blocks of service efficiency and effectiveness by establishing procedures, which reduce the segregation within and between the processes inherent to transit service deployment.

Run-time, layover requirements

Once the importance of avoiding segregation is accepted, it becomes clear that the service plan must find balance between service effectiveness and operational costs. In concrete terms, what this means is that planned route length and frequencies must yield efficient Round Trip Cycle Times (RTCT).

RTCT is defined as the time required to undertake a round trip, including both runtime and mandatory layover time. Understandably, route runtime and service frequency will vary considerably throughout the day, as will layover time — mandatory to address runtime volatility. How then, can we hope to plan efficient RTCT for every period of the day? The answer to this will come in due time. For now, let us elaborate on RTCT efficiency by discussing run time and layover calibration, the incumbents of punctuality.

The development of accurate run times and layover requirements, using a methodology, which will promote service punctuality in line with the agency’s policies, will be the foundation of efficiency. With accurate runtimes and layovers requirements, the blocking and run-cutting processes will be permitted maximum efficiency and not require the inclusion of inefficient cushions to protect inaccurate schedules.

There are several well-documented techniques involved in runtime and layover calibration; while we cannot elaborate on each, it should be clear that expertise and software solutions are widely available. The only challenge, which will have been addressed by the SEA, is the identification of which techniques and/or tools will work best for a given transit system based on its operating environment.

Once runtimes and layover requirements have been established, the RTCT efficiency can easily be assessed using a standard spreadsheet.

An efficient service plan is one which will reduce excessive cycle time and limit vehicle requirements even under the pessimistic worst-case scenario of stand-alone blocking.

 [PAGEBREAK] 

Cycle-grouping

As mentioned earlier, fluctuations in runtimes, layover requirements and service frequency can render RTCT efficiency extremely challenging. A highly valuable approach to this problem consists on a time-tabling approach that groups inefficient RTCT so as to cumulate efficient cumulated cycle times. Let’s call this cycle-grouping.

It is important to understand the clear distinction between interlining and cycle-grouping. Interlining occurs within the blocking process and is not planned, it is opportunistic. Interlines have significant benefits in the peaks, such as reducing peak vehicle requirements, but they also have significant drawbacks such as tying the entire system into a highly volatile domino board. If one route suffers operational challenges, the entire transit system is affected.

Cycle-grouping is different. It is a form of interlining that obeys the following principals: first, it is established during the time-tabling process. Cycle-grouping involves the grouping of routes, typically two and seldom more than three, which will be synchronized so that the departures of one route will synchronize with the arrivals of another. Groups will be methodically chosen (as opposed to random or opportunistic interlines) so as to systematically cumulate an overall efficient RTCT while giving consideration to passenger transfer information, common trunk inter-spacing, vehicle type preferences and terminal capacities. Because of the grouping, if one route fails operationally, no more than two routes are affected —­­ the operational crisis is contained.

Blocking efficiency

Thus far we have ensured accurate runtimes, documented layover requirements and developed timetables that circumvent as many planning inefficiencies as possible. Whether we use manual or so-called “automated” blocking, we must ensure we respect the efforts of our predecessors and assist as best we can to improve what was beyond their leeway.

There are several contributors to scheduling efficiency within the blocking phase: trip shifting to flatten peak vehicle requirements, alternate vehicle type preferences to optimize fleet and route-garage allocation for load balancing across multiple divisions.

Amongst the less frequent or under-utilized blocking efficiency opportunities, we typically find the development of an accurate system-wide deadhead matrix to ensure all blocking opportunities are explored and the adjustment of blocks to promote runcut efficiency.

Run cutting efficiency

Computer-assisted scheduling is amongst the most significant contributors to run cutting efficiencies. As complex as they may be, every labor agreement can be translated into a clear set of rules with which computer systems can deal with unambiguously. As computers are faster every day, computing and correctly evaluating millions of possibilities within minutes is standard.

Computerized scheduling systems allow users to reduce the trade-off gap between the time taken to produce scheduling solutions and the quality (or cost) of those solutions.

Another aspect of run cutting efficiency is to delineate between objective criteria, which mathematical models can deal with in a highly effective manner, and subjective criteria, which are more difficult to model and often require the user to work interactively. A potent tool must allow the user to be able to accurately model objective constraints and provides significant tools to consider the results of algorithmic runs and manually address subjective criteria/goals.

Other examples include software calibration following the assessment of fringe benefits to balance the cost of overtime with that of hiring an extra operator. Somewhat more complex, the run cutting practice of integrated scheduling is gaining momentum.

Integrated scheduling essentially consists of enabling the run cutting function to undertake small blocking changes to improve run cutting efficiency without adverse impacts on blocking efficiency. The concept behind it is that there are many alternate blocking solutions, which are cost-neutral from a vehicle-miles and vehicle-count perspective, but not all of which will ease and/or enable comparable run cutting efficiencies. Integrated scheduling typically requires a powerful scheduling tool, as well as staff fluent in the most advanced run cutting and blocking concepts. For this reason, implementation of integrated scheduling requires significant investments in staff training, software tools and time.

Agency-documented efficiency gains resulting from SEA projects have ranged from 1 percent to 3.6 percent of annualized operator hours. When runtime and layover calibration was part of the SEA, highest gains are observed with documented on-time performance improvements as large as 14 percent.

Experience has shown that roughly half of the efficiencies are directly related to cycle-grouping and elimination of the segregation, while the other half involved revised manual and/or automated blocking and run cutting techniques.

Michel Courval is a consultant with a focus on public transportation planning and scheduling. He has worked with dozens of private and public transit agencies worldwide developing and implementing manual and computer-assisted scheduling practices for the production of schedules and rosters.