Traditional financial tools do not allow airlines to correct inefficiencies as they arise, a fault which can be rectified by the newly developed technique of process controlling. Report by Wendy Nichols and Harald Deprosse. It could have been any airline at any airport. The head of the check-in department was taking his last look around in the early shift. It was the first day in the holiday season and, despite the fact that such days are notoriously busy, he was once again ill-prepared for the onslaught of passengers and the never-ending queues. Indeed, this time, it had been so bad that he had even had to check-in passengers himself at 7:30 this morning.

Major airlines are finding themselves increasingly stuck in a rut: tumbling yields are forcing them to slash costs even as service quality becomes more important for product differentiation. Nowhere is the dilemma perhaps most acute than on the ground, where endless queues at the check-in counters often lead to a degree of passenger irritation that no amount of champagne and strawberries on board can soothe away. And, with ground services already absorbing 10 per cent of an airline's operating costs (excluding landing fees), it is clear that throwing money at the problem is no solution.

Yet experience with some European airlines reveals that - even taking into account existing legal and union restrictions - productivity increases of 20-40 per cent in various passenger ground service functions are possible without a deterioration in quality standards. The key lies not so much in speeding up the service procedures themselves as in eliminating the 'deadtime' caused by fluctuations in demand.

A typical check-in procedure with luggage check takes less than two minutes from start to finish. Yet staffing levels at European airports typically allow for 5 minutes of active working time per passenger. When holidays and other days off are factored in, airlines are actually paying for a staggering 6-7 minutes check-in time per passenger. In productivity terms, this implies that the A-league of European carriers check in between 16,000 and 20,000 passengers per operational employee per year. Taking into account local wage rates, this means labour costs of US$2.80-US$3.50 per passenger. 'It looks to me like they should have enough time to escort each passenger personally to the gate,' observes one airline manager. However these comparisons should be treated with caution given that check-in procedures vary between airports.

Why is it then, that current attempts to cut the flab have led to even longer queues, distraught employees and (in some cases) escalating sickness rates? A closer inspection of the planning processes for passenger ground services indicates that airlines are failing to come to grips with fluctuations in demand whether they are seasonal, weekly, or at certain times of the day. And, where adequate planning exists, it is often not adhered to.

All too frequently, there are no advance warning systems in place which can recognise under or overcapacity before it appears. Classic financial control instruments catch sight of the rising unit costs and the all-too-frequent budget overruns, but are powerless to indicate the causes. The same applies to service quality control systems, which can only recognise the symptoms, not the disease.

Simultaneously improving productivity and service quality in ground services calls for matching staffing levels more closely to the peaks and troughs of demand throughout the year. This will both improve the unit cost position - personnel costs typically make up 75 to 80 per cent of the cost of providing ground services - and service quality, while contributing to a less stressful environment for the employees. Essentially, the goal must be a constant productivity irrespective of seasonality. While the goal is clear, the main difficulty is to establish an effective reporting and controlling system to accompany the new planning process and enable the continuous correction of deviations from plan. This in turn requires effective planning instruments; planning, it has been said, is the substitution of chaos with error, and correcting those errors before they materialise is the business of process controlling.

Process controlling is the generic term for an innovative method of capturing non-financial performance data to ensure the optimum outcome. In the case of passenger ground services this involves: breaking up planning processes into their constituent steps; establishing a system of key performance indicators and ratios; and building up a tool-kit to correct deviations.

For check-in procedures, the personnel capacity planning and allocation process is split up into five steps:

1 Capacity planning or the number of full-time and part-time employees needed in the budget planning period;

2 Shift planning - over how many weeks and with what shift lengths;

3 Absenteeism planning or the number of days of absence and when should they be granted;

4 Roster planning - is the number of employees needed adapted to the latest forecasts?

5 Deployment or how the allocation of employees can be event-related.

For each of these five steps there are a series of key performance indicators (see diagram 1). Roster planning, for example, depends for its success or failure on the accuracy of workload forecasts (expressed in required manhours) and the precise coverage of that workload.

The key lever is the control of planned absences such as holidays and vocational training. If only half of the available workforce is needed on a given day, the other half ought to be 'planned absent.' Failure to control this lever properly can lead to productivity fluctuations of up to plus or minus 30 per cent in the course of a year, a sure-fire recipe for dissatisfied passengers and frustrated employees.

The process controlling system compares, long in advance, the correlation of employee days off with the expected workload, and flags any deviations. This reporting measure is varied dynamically with changing forecasts of workload, which are in turn based on yield management's analyses of booking behaviour.

The degree to which deviations occur enables departmental planners to be kept informed of the implications of new forecasts on their activities so that they can make more capacity available through shifting training and vacation days, or reduce capacity by doing the reverse. An exact match is only attainable through union agreements which allow the airline some measure of flexibility - especially difficult for European carriers that deal with strong unions and strict labour laws. One apparently successful model allows employees to plan 80 per cent of their days off freely, leaving the remaining 20 per cent subject to flexible planning once the most recent forecasts are available.

The essence of a successful process controlling mechanism is its informational content. No longer are planners and departmental heads left in the dark regarding their operational effectiveness. By flagging the deviations in advance of the day in question, process controlling gives planners the opportunity to intervene, and more importantly, tells them what they need to do to achieve their productivity targets. Diagram 2 illustrates how the interdependency between one process step and those that follow can enable control to be regained over a planning process.

An inherent learning effect is also integral to the planning process. Persistent, lingering deviations from plan, for example, can indicate where forecasting tools have to be improved.

Ultimately, process controlling makes the performance of managers in ground services more transparent. In a traditional financial control environment, the responsibility for bad results is usually shared by all the participants in the planning process. When managers can be held accountable, performance-based remuneration becomes more practicable.

The benefits of process oriented control systems can also be reaped in other personnel-intensive areas of an airline, such as ticket sales, ramp handling, flight operations and maintenance. Process controlling can help managers get a grip on costs and quality wherever long planning periods coincide with strongly fluctuating workloads and high unpredictability and instability.

Implementing process controlling involves five basic steps. The first step is to identify the proper performance measures. Process controlling generally goes hand in hand with the general reassessment of a business process. In the course of 'redesigning' a process, the key performance indicators for each process step reveal themselves almost automatically. It is also critical to identify interdependencies between the various performance indicators. Just as financial controls break the return on investment down into a hierarchy of constituent measures, process controlling seeks to identify the 'elementary particles' of process effectiveness.

The second step is to ensure the availability of the relevant data. An airline that does not monitor statistics such as planned versus actual manhours worked, or planned versus actual days of absence, cannot fully benefit from process controlling.

Thirdly, the responsibility for gathering data and tracking the performance indicators has to be defined, taking care to separate the evaluators from the evaluated. And fourthly, procedures should be established for taking action when something goes wrong. Errors in earlier planning stages must be corrected at a later stage ('feedforward') and prevented from recurring ('feedback').

Finally, the responsibility for the outcome must be firmly assigned to the designated decision makers at every level. Shift planners, for example, are held responsible for the proportion of 'deadtime'. Heads of departments require such indicators as productivity per manhour or average waiting time per passenger. Airline station heads require more aggregated indicators such as passenger satisfaction with the entire service, manpower costs per checked-in passenger, and so on.

The cost of establishing and implementing a process controlling system varies according to the existing information base of the airline. For those with more sophisticated IT systems, a new process controlling system may require little more than a few additional programming hours. The evaluation and presentation of the performance indicators then takes up a few hours a week.

The situation is more problematic for those airlines with a poor information base. In this instance gathering data can be a menial task, such as counting the various days of absence in a shift plan. A short term solution is a random sample taken by intensifying observations made over a period of a few weeks. But in order to reap the full benefits of process controlling, investment in a state of the art IT system is unavoidable.

One problem for any control system is reflected in the truism, 'What gets measured, gets done.' Permanent controls are needed to ensure that quantitative measures don't become goals for their own sake. The implementation of a process controlling system may meet with resistance from the various employees whose performance now becomes measurable. It is therefore important that the chosen ratios and target values included in the system are agreed with the employees.

A further major challenge is to inspire the cultural change that has to accompany the new management goal system. The aim is to get employees to internalise the documented process goals - not just to practice process orientation, but to live it. If the airline succeeds at that, process controlling can be transformed from an instrument for reporting eternal shortcomings into a tool for inbuilt competitive advantage.

Source: Airline Business