At the 1996 Conference, SC1 and SC4 were asked to include the Productivity and Performance Indicators onto their working programmes. At the Taipei conference, in March 1997, SC1 presented a working paper containing definitions of performance indicators. SC4 indicated that they had the greatest of difficulty in addressing the subject, because of the various interpretations that could be made on the subject of productivity. A working paper on productivity was presented, which gave a brief summary of the work SC4 had done so far.

The subject remains very complex and difficult as the term ‘productivity’ is mainly used in commercial industry and usually refers to manufacturing output. When talking about the productivity of a service, only a few publications offer information about how productivity can be defined in this respect.

As it is not easy to find a proper definition for productivity in regard to the provision of an ATC service, this working paper will give an explanation of what productivity means when applied to commercial industries. As an example, the car manufacturing industry will be used and the conclusions reached will provide a basis for our own interpretation of how productivity should be defined for the ATC service.

Airline companies are interested in keeping their costs as low as possible and it is natural for them to urge ATS services to lower their service fees, whilst at the same time seek improvements in the provision of the service. In a commercialised world everyone is talking about productivity and the need to reduce costs. Many of the ATS organisations give the impression that their objectives are to enhance productivity, without actually defining what they are actually doing to achieve this. Privatised ATS organisations have a big interest in lowering their charges, in order to be capable of competing on the market and to maximise profits. To be able to offer ATC services cheaper, they say that productivity in ATC has to increase. In addition, ATS companies are defining performance indicators in order to determine the productivity of particular ATS divisions. In some countries, these performance indicators are already used to define the complexity of work in certain ATS divisions. ATCOs within the same country sometimes receive different salaries depending on the specific ATS division they work in. Often, no research has been made to obtain a form of scientific background to justify these indicators. As both items have, or will have, a great influence on the work of an ATCO, it is important to define IFATCA’s policy on the interpretation of Productivity and the correct use of performance indicators.

Productivity

Definitions

In commercial industries there are two main terms when talking about productivity. One is productivity in general and the other one is the so called ‘working productivity’. All of the following definitions can be found in various dictionaries of economy (sources for this paper see Reference Documentation at end).

The general definition for productivity reads:

“Productivity is defined as the relation between output (e.g. number of produced cars) and in put (e.g. amount of steel which is used to produce these cars)”.

In respect to working productivity there are two ways of defining that term:

“Working productivity is defined as the relation between the number of produced items (e.g. number of produced cars again = output) and the number of workers respectively the number of in vested working hours (input)”.

“Working productivity is defined as the relation between the annual turnover (output) and the number of workers respectively the number of in vested working hours (input)”. (MühIbrandt, Wirtschaftslexikon (Dictionary of Economy), 1996)

Output and Input

In respect to commercial industries the definition of output and input values is quite easy. Examples for each can be found in the above mentioned definitions. If we look at productivity in ATC from this commercial point of view, similar values could be laid down. The following enumeration will give some examples.

Output can be:

• an increase in the annual turnover;
• an increase in number of aircraft to be handled within a certain time;
• lower fees for airline companies;
• low rate of incidents / operational errors.

Input can be:

• number of ATC personnel (whether higher or lower);
• number of working hours (whether more or less);
• sectorisation (e.g. the combination of two sectors into one new sector); hardware (better and/or newer systems);
• automation (e.g. automated revisions);
• procedural changes (e.g. reduced separation minima).

An example of ‘input’ values:

The productivity of an approach sector could be increased by the implementation of reduced separation minima on final approach. That would enable the controllers of that sector to increase the number of aircraft landing within the same time period as before.

Productivity in commercial industries

The productivity of a single worker at a conveyor belt for example could be defined as relation between the number of cars (output) he is working on within his time of working hours (input).

To increase the productivity of this worker there are two possibilities:

• His working time of the employee could be reduced without reducing the number of cars he has to handle. So his productivity has increased because he is working on the same number of cars within less time.
• The second possibility is to let the conveyor belt run faster so that the worker is forced to handle more cars within the same time period.

The highest degree of productivity increase in this example would be to combine both possibilities.

It is easy to imagine that a conveyor belt running faster would result in an increase of the overall productivity of the total company no matter whether we talk about productivity in general or about working productivity. However, these actions can also result in poorly assembled cars and raised personal injury rates.

Productivity in ATC

As will be shown in this paragraph, it is nearly impossible to find any proper definition for the term productivity in respect to ATC. The question has to be asked whether it is useful at all to talk about ATC-productivity as the main task for any ATCO is to “promote safety, efficiency, and regularity in international air navigation.” (IFATCA, a vision document, 1997) This is the main objective of IFATCA as it is laid down in the IFATCA mission statement. So we additionally encounter the problem of having found a new kind of productivity. Let’s call it Safety Productivity. An increase in productivity then would mean an increase in safety. This would be measured by maintaining a ratio of ATC related occurrences against the number of aircraft moving through the unit’s airspace. It would however, not fit into any productivity model studied by SC4.

As ATC do not produce goods, productivity in this respect could probably be defined as working productivity in the sense of the definitions above regardless of whether the ATS organisation has been privatised or is a governmental institution. It is accepted that ATS organisations are willing to embrace an increase in productivity and it is important to employer and employee that the correct understanding of productivity is determined. Productivity as a relation between the annual turnover and the number of invested working hours – as said in the definition – would normally consider each employee of the company and thus is not only an ATC problem.

An increase in ATC productivity could as well mean to be capable of handling more traffic for instance in a certain sector. As this increase might have an effect on air traffic safety, it is very important to find reliable criteria by means of which it is possible to state what the present level of productivity is. An increase in productivity in this context must never lead to overload or unsafe situations. Even if that is covered there are still some more problems. The capability of controllers to handle a certain amount of traffic differs from day to day, and within a day, for various reasons (e.g.: traffic situation, complexity of traffic, partial or complete system failure, private problems, fatigue and a lot more). Another problem: Imagine, that the ATCOs of a certain facility have been trained to be able to handle more traffic. What if there simply is not more traffic than before? Productivity would have increased but from an outside point of view no-one would recognise it.

Counting only the number of aircraft per hour or kilometres flown within a certain sector or facility would not take into consideration the complexity of air traffic. Additionally it is obvious that we cannot say that a certain sector is the more productive the more complex the traffic within it is. A controller will be able to handle more traffic in a less complex situation, but being able to handle only a few aircraft in a situation which is very complex does not automatically mean a reduction in productivity. The capability of a controller to cope with different degrees of complexity is highly dependent from numerous factors like e.g. the technical equipment the controller is forced to work with, relief times, staffing numbers, the personal ability of the single controller, his or her degree of education, personal situation and so on. Additionally every controller has got his or her own opinion about which situation is complex and which is not.

As another factor, overall traffic workload has to be considered. According to the definitions of productivity any change in workload as an input factor would mean a change in productivity. This is not quite the same as mentioned about complexity. The overall workload of a certain sector might be very high when counting the number of flights only, but the traffic situations within this sector might not be very complex. Of course, it could as well be the other way around. Overall workload is a very important factor to the ability of a controller to work safely as there are a lot of limiting factors (exhaustion, limitations of the equipment, staffing number and so on).

The next main point is that there still exist a lot of other outside factors which have an effect on an ATCO’s work. Some of these factors are: weather situation at an airport which is served by the facility, flow control measures at other ATC facilities, limitations of airports, technical restrictions, lack of personnel due to different reasons and a lot more. As an additional example we can take a look at the report, “Delays to Air Transport in Europe, January 1997” which was published by the Central Office for Delay Analysis of Eurocontrol. This report is talking about the reasons for the significant increase in the number of delayed flights in comparison to January 1996 (total increase in delay of about 50%!). The main reasons for this increase were: “adverse weather situations; aircraft incidents and equipment failures, staff shortages, parking problems at a certain airport, industrial actions and continuing congestion due to runway re-configuration at a certain airport”. ATC-based delay was a factor as well but showed up with 18% only. 82% of the delays were caused by other influences.

It is very important to realise that productivity in ATC is very different from productivity in commercial industries. In ATC no produced goods can be found which could be counted to say that a certain facility is working more productively than another. On the other hand, traffic numbers are often used as the only item to measure productivity, not taking into consideration complexity and workload. In fact, productivity in ATC is a conglomeration of a lot of different factors which have great influence on the output.

Complexity of air traffic and workload might be the most important items but there are other factors (as shown above already) which are not as obvious as these two. Obviously it is better and more correct to talk about complexity and workload instead of productivity. As complexity and workload have a great effect on air traffic safety, ATC organisations and ATCOs should find reliable criteria for these two items first, taking into consideration all possible factors. Figures have to be found to state up to which extent a complex situation is still workable without affecting safety. Peak workload situations have to be considered as well as technical limitations, staff limitations and all other possible influencing factors. Any proposals to increase workload productivity must not have a detrimental effect on safety productivity. The main point for any ATS organisation should be to improve or to hold the level of safety in respect to air traffic.

Performance Indicators

A lot of ATS organisations use some kind of indicators to measure the ‘productivity’ of a certain ATC unit, sector or of an ATCO. Additionally, the overall workload of a certain unit or sector might be estimated with the help of special indicators. Even safety can be ‘measured’ by counting the number of ATC related incidents or near misses for example.

Performance indicators can be useful for controllers to identify when traffic increases, to identify required staff levels and system and/or controller workload capacity. Any such performance indicators must be taken as a mean, to compensate for peak traffic workload fluctuations, extraneous influences (such as weather factors, delays, holding, diversions, system reliability, equipment malfunction, average staffing levels etc.).

In addition, it has to be accepted that as a human, a controller has a finite workload ability. When determining workload abilities for all controllers, an average value must be established to reflect individual aspects, as there are : “experience, workload, stress and fatigue, boredom, attitudes, trust and job satisfaction.” (V. David Hopkin, Human Factors in Air Traffic Control, 1995)

It is important to scientifically measure and determine the peak capacity of any airspace and, taking into account the “average controller’s ability” to handle traffic, determine mean values for workload and performance related to productivity payments. Regardless of any performance related productivity measurements, it is essential to identify and limit the amount of traffic that controllers can work safely.

In some countries, catalogues of different indicators exist by means of which the salaries for the ATCOs of different ATC units are calculated. No single performance indicator can adequately measure productivity of any ATC sector. Therefore it is important that performance indicators are not used as stand-alone items. Only the use of a number of them might give an overview of complexity within a certain region. A single indicator like ‘number of flights per hour’ does not consider airspace structure, complexity of traffic, performance of aircraft and so on.

The most important task of any ATS service is to provide a safe air traffic environment. IFATCA could not define ATC productivity or identify a single performance indicator that could adequately measure overall ATC productivity.

It is recommended to Conference that this paper be accepted as guidance material.

Wirtschaftslexikon (Dictionary of Economy (translation by author)), F.W. Mühlbrandt, Berlin 1996, ISBN 3-589-21081-8.

Die neue Suche nach Spitzenleistungen (orig.: What America does right), R.Waterman, Düsseldorf 1994, ISBN 3-612-26239-4.

ATC Performance Indicators by Bert Ruitenberg, found in the AVSIG forum of CompuServe

Air Traffic Controller Awareness of Operational Error Development by M. Rodgers, Ph.D., FAA, 1993.

Human Factors in Air Traffic Control, V. David Hopkin, London 1995, ISBN 0-7484-0357-4 (paperback).

Delays to Air Transport in Europe, January 1997, Central Office for Delay Analysis, Eurocontrol, Brussels.

IFATCA, a vision document, IFATCA 1997.