This paper was presented at TRIZCON2002, The Altshuller Institute meeting in St. Louis, MO, USA, April 2002 and published in the proceedings of that meeting.
TRIZ Introduction In Airlines Airport Management Division
The SANNO Institute of Management
In order to realize an accident-free situation, the airline industry has made various efforts. Recently, the industry followed the approach from the aspects of psychology and behavioral science aiming at reduction of human errors. This approach, however, has produced results in terms of occurrence prevention, but is not enough from a perspective of recurrence prevention. Introducing TRIZ here helps the airline industry to realize an accident prevention measure, including an entire flight system, in order to establish a system for analyzing causes of an accident and preventing their recurrence. This paper will discuss the case where the TRIZ methodology was applied to human error prevention training in the airport management division of Airline A.
1. overview of the airline industry and their current approach to preventing aircraft accidents
1.1. Current Situation of the Airline Industry
The number of commercial aircraft flights*1 in 2000 worldwide was 21.20 million with a total flight time of 38.20 million hours, increased by 11.5% and 8.3% respectively from the previous year. The number of passengers transported was 2,017 million, a 6.6% increase. Although there is a temporary reduction due to special circumstances, it is anticipated that this upward trend will not change in the future. On the other hand, the direct damages by accidents*2 are assumed to be $550 million, and the total amount of damages including compensations to passengers and damages by small accidents to be $1,668 million. It is needless to say that prevention of accidents not only leads to a sales increase due to improvement of reliability from the aspect of safety, but also is one of the major management challenges due to its serious influence of accidents on financial items.
1.2. Overview of Aircraft Accidents
The history of aircraft consists of struggles in accident prevention. Even now the world airline companies make various efforts to prevent occurrence of accidents.
During the period of 43 years from 1958, the flights of jet airplanes made numbered 393 million for 634 million hours in total, with 800 total-loss accidents having occurred. Among these, 630 accidents occurred during flight. If this is analyzed by the rate of total-loss accidents per 1 million flights, the number of accidents decreases: 5.28 accidents for 43 years; 1.34 for the past 20 years; 1.27 for the past 10 years; and 1.11 for the past five years.
This dramatic improvement greatly owes to performance improvement of the aircraft body and development of radio navigation. At present, it is not possible to say that the accidents during flights can be eliminated, but the major causes of the accidents have been changed. IATA Safety Report (JET) 2000 indicates that the causes of the current accidents are  human error factors (30%),  organizational factors (25%),  environmental (weather) factors (24%), and  technical factors (20%). Accordingly the actions to prevent accidents have been changing from those in the technical aspect of aircraft and navigation to those against human error and organizational factors.
1.3. What are Human Errors?
M.S. Sanders and E.J. McCormick, US ergonomists, defined human error as �an inappropriate or undesirable human decision or behavior that reduces, or has the potential for reducing, effectiveness, safety, or system performance.� There are other several significant definitions, and thus it can simply say �a general term for errors caused by human decisions and behaviors.�
Figure 1: Points of Generating Errors by Human Factors
The important point here is the fact that human decisions and behaviors are constantly influenced by natural characteristics of human beings as a living thing as well as tools and machinery, facilities, environment and other people involved in the course of conducting their jobs. From this perspective, so far the prevention of human errors has been approached mainly by understanding human�s natural characteristics -- the core of the system constructing jobs -- from the perspectives of psychology and behavior science. The focus here is to understand the mechanism of an error caused by human factors in the sequence from human�s input = perception/recognition; intermediate process = judgment/decision-making; to the output process = operation, as shown in Figure 1, and not to generate such an error.
1.4. Approach to Human Error Prevention
The airline industry has already developed the CRM (Crew Resource Management) training program, which is a training approach mainly from the perspectives of physiology and behavior science and engaged in reduction of human errors by pilots. However, in order to improve safety of current operations supported by a quite complicated system, it was recognized that the training only of pilots was not enough and the US Federal Aviation Administration (FAA) decided to mandate the CRM training in 1995 as well as expanding the applicable personnel for the training (e.g., dispatchers and cabin attendants).
In Japan the Civil Aviation Bureau of the Ministry of Land, Infrastructure and Transport also established the legislation for the CRM training. At present, the airlines have gradually expanded the training scope to cover dispatchers and cabin attendants in addition to pilots.
1.5. Improvement of Conventional Approach
The conventional approach to the prevention of human errors from the perspectives of psychology and behavior science worked very well on reduction of aircraft accidents. This approach, however, is not sufficient from the two points.
First of all, an approach from psychology and behavior science is reasonable from the viewpoint of occurrence prevention, but it tends to put too much emphasis on mentality, thus leaving doubt in terms of recurrence prevention. In other words, the experience of an accident accompanying huge sacrifice can be capitalized only as recognizing anew the significance of human errors.
This is because an accident caused by technical factors of aircraft often left physical evidences, but for human errors although it is possible to assume behavior condition, it does not have accuracy as firm physical evidences. What�s more, the reason why the person took that action causing the accident cannot be explained easily.
In order to determine that human error caused the accident, it is necessary to prove that proactively. But in reality while the cause is still vague, it is determined that the accident was caused by the human error just because a human was there. This kind of ambiguous treatment makes capitalization and recurrence prevention difficult.
Secondly, as the above-mentioned IATA Report stated, there are four major factors, which hinder safety. But as only human errors are treated as an independent object from other three factors, the linkage as a whole systems is weak. Investigation reports of aircraft accidents well demonstrate the difficulty of identifying one cause of the accident. In most accidents, factors are overlapped and it is quite difficult to determine which is a main cause and which is a secondary cause. For example, if any mechanical equipment failed first and then a pilot could not take any action against it, which is a main cause? Originally, whichever the main cause or secondary cause, action must be taken. But because the relationship between the human error and accident is stressed too much, this approach has potential risk that the cause of the accident is distorted as simply a problem of the individual and disturbs the progress of recurrence prevention measures, which are the essential objective of the investigation of the accident.
F.H. Hawkins of KLM Royal Dutch Airlines, who established the basis of the CRM training, already explained the importance of the entirety mainly the mutual relationships among the factors using the SHELL model. However, although the SHELL model is excellent in terms of explaining the importance of the entirety, it leans relatively to the pilot position and lacks concreteness from the perspective of the whole organization.
Consequently, connecting the current approach mainly using psychology and behavior science to the entire system naturally is an important perspective for preventing accidents through prevention of human errors.
Figure 2: Structure of Factors related to Accident Prevention
Figure 2 is the Structure of Factors for Accident Prevention prepared by Tadashi Watanabe*3 and modified by the author. In this illustration, the bold frames are the elements of the approaches to human errors in the conventional CRM training program, and when considering accident prevention, this illustration clearly shows that they are not independent. There are factors related to technology, procedures and organizational structure, and together with human error factors, they are entangled as the main and secondary causes. Thus, it is quite obvious that all the factors should be examined together and measures to prevent accidents including the linkage of the factors should be taken in the future accident prevention.
2. TRIZ application to program development
2.1. Airport Management Division of Company A
The airport management division provides support to aircraft operations and performs all the handling services. Especially the station control, in order to realize arrival and departure in accordance with the schedule, must determine airplane parking locations, control weight of passenger cargo, provide information on the aircraft body and the latest route, supply fuel based on a flight plan approved by the pilot, instruct the passengers boarding and alighting, check the maintenance situation and adjust the time, upon coordination with a dispatcher as well as the passenger, cargo and maintenance divisions, within the time from arrival to departure of the aircraft (maximum: about 50 minutes for B747 and minimum: 35 minutes for A320). What�s more, as arrival time tends to be delayed in the actual operations, this airport management division requires making efforts to shorten the work time by even just one minute and realize the arrival and departure exactly based on the schedule.
They must perform extremely diversified and complicated operations in a short time, and therefore, although systematization and automation have been progressing, human errors are not irrelevant. Especially, decisions are made mainly based on the information obtained via radio, and a possibility of occurring perception/recognition errors and judgment/decision errors always exists. Thus, for the past few years, the division has made efforts to prevent human errors through participation in training for pilots and dispatchers.
However, as human errors by the airport management operations have less severe impact on direct safety compared with those by pilots and dispatchers, the tendency of recurrence of same errors is high and especially a necessity of perspective of recurrence prevention was perceived. The operations of this division have a characteristic that most of their operations directly reflect on customer satisfaction such as punctuality, efficiency and comfort. Thus it was recognized that the training program suitable for the operational characteristics of the airport management division is necessary.
2.2. Motive for TRIZ Application
For the airline operations, not only safety but also punctuality, if customer service is considered, are required. But the operational level has a strong belief that realization of safety and that of punctuality are opposing. Actually under the time-pressing condition, humans tend to make mistakes in decision-making. From this kind of experience, the people of this division have the psychological inertia -- safety and punctuality are confronting parameters, so we do not spare any efforts, but ideal solution for realizing both never exist.
Thus, we made the people related to this division understand that, if the past specific operational improvements are analyzed with the TRIZ concepts, many operational improvements can be explained using TRIZ inventive principles and the problem of safety and the other contradicting parameters can also be solved idealistically in the future. In this way, release from the conventional psychological inertia was promoted.
Some criticize that these 40 Inventive Principles and Separation Principles are difficult to apply to non-technical fields. But as they can eliminate a psychological inertia, which disturbs pursuing ideality, if an expert who can adequately convert the cases uses them, they can work as an effective tool.
For example, if the problem is viewed in the opposing relationship of [Accuracy - Time], four perspectives of thinking can be obtained. The highest priority principle, [24. Mediator], has the sub-principle of [a. Use an intermediary object to transfer or carry out an action]. The improvement of the ground handling method currently undertaken by Company A is an attempt to improve work accuracy and time accuracy by appointing a manager between the station control division and the work related to aircraft. This can be explained using the 40 Inventive Principles.
Figure 3: Contradiction Table and Modified 40 Inventive Principles
Application of TRIZ to several cases, which relieved psychological inertia from the project members for developing the training program, as a result, proved the improvement of reliability of TRIZ as a problem solving tool and effectiveness as a tool for eliminating psychological inertia of participants of the training program seminar. Then it was decided to use TRIZ as a basic tool for problem solving in the program.
2.3. Motive for AFD Application
In order to prevent recurrence of human errors of pilots, the safety reporting system has been established in the advanced aviation countries. This is a system to accumulate human errors that actually occurred and make use of them for prevention in the future.
As mentioned above, errors occurring in the airport management division have less serious direct effect on safety, so the same errors were repeated, and thus it was recognized that the perspective not to repeat the same error is further important. What�s more as the airport management operations have many related divisions and the quality of errors is different from the navigation operations. Therefore, it was determined that for recurrence prevention, more accurate analysis and development of solutions are required. Thus, it was reviewed as to whether the AFD (Anticipatory Failure Determination) developed by Boris Zlotin et al. can be used as a tool.
In the process of this program development, the operational functions of the airport development division were defined. As a result of reviewing what kind of method can be considered to deteriorate these functions, in most cases, satisfactory results could be obtained, and it was decided to use AFD as a tool for error analysis and recurrence prevention.
It was also concluded that as an analysis using AFD handles all the elements (both useful and harmful elements) drawn in the Problem Formulator equally without any prejudice, analysis while maintaining the entirety of the system was possible.
2.4. Problem when Introducing TRIZ
It is well known that TRIZ is effective for suggesting viewpoints of grasping a problem and a direction for solution not only in the technical field but also non-technical field. However, in order to provide it as a tool in the airport management operations (non-technical field), the same way as when providing it to engineers is not sufficient. Thus, in this program, within the range where the TRIZ essential excellence is not impaired, modification where the non-engineers can accept it without any confusion was required.
The essential excellence of TRIZ is, as a matter of fact, the �Ideality� concept, and making people understand how the problem should be viewed and providing what kind of perspective of thinking can be used to generate ideas for solutions, including effective use of resources.
3. Direction of TRIZ modification
3.1. Overview of Modification
Introduction of the Classical TRIZ techniques was difficult because the case is non-technical, had a time limitation together with its volume. Thus the limited techniques from below (1) to (4) were selected and introduced after modifying them to make them suitable for the actual operational characteristics. Modification here means mainly expressing the Inventive Principles in an easy to understand way for non-engineers as well as limiting the techniques to use.
3.2. Way of Providing Contradiction
In TRIZ understanding of Contradictions is important in learning basic perspective of problem analysis. But it is also a fact that even if 39 Engineering Parameters and 40 Inventive Principles, which construct the Contradiction Matrix, and Separation Principles are explained as they are, they are too confused to use for non-engineers. Thus, although addition of non-technical explanation is required, for application of the Contradiction Matrix in a non-technical field, as 39 Engineering Parameters are not necessary, but just a small number of them is good enough. So the descriptions of 19 Parameters were created for non-engineers. 19 Parameters are Speed, Stability of object, Strength, Power, Loss of information, Waste of time, Reliability, Accuracy of measurement, Accuracy of manufacturing, Harmful factors acting on object, Harmful side effects, Manufacturability, Convenience of use, Repairability, Adaptability, Complexity of device, Complexity of control, Level of automation and Productivity.
For the 40 Inventive Principles, the defect that the sentences are slightly longer was ignored, and describing on the level of sub-principles was used. This makes it easier for non-engineers to understand these principles well, then descriptions were modified. As a result, easiness for using the principles in non-technical field was improved substantially.
Physical Contradictions can be applied to non-technical field more easily than Technical Contradictions. The Separation Principles have a quite high abstraction, thus applying to non-technical field is easy.
3.3. Problem Formulator to Accurately Grasp the Problem
The method to draw a complicated problem actually occurred in Problem Formulator by adding the drawing method of Function Information Analysis while maintaining the entirety of the problem was developed and introduced into the program.
The Function Information Analysis is a technique to analyze operations by clearly dividing functions, information and files. The objective of introducing this method is to grasp the problem situation definitely while maintaining the entirety. Once the problem situation is defined, the perspective for solution can be obtained through application of Operators.
3.3.1. How to Draw a Function Information Analysis Type Problem Formulator
The basic rule is to draw one by using two types of elements related to the problem -- functions and information -- after dividing them into useful or harmful. Then, the mutual relationship between the functions are linked using the four types of arrows.
As a result of a review actually using this technique, focusing on things, which seem to be the causes of the problem, could be performed more clearly than the analysis using other techniques. For example, [Information was not delivered] became a more appropriate analysis result [Information was not delivered in a timely manner]. Then for the idea of preventing recurrence, the direction of eliminating [Something disturbing the timely manner] could be obtained.
Figure 4: Symbols Used and their Meanings
|Function||Element as a function in the system. Can be expressed as �to do something.� Black is useful and red is harmful. The organization name is entered in the lower section.|
|Information||Element as information in the system. Can be expressed as �to instruct or to deliver.� Black is useful and red is harmful.|
|File||Files and various regulations, which are the basis for judgment.|
|Supplies Causes||Black arrows show supply from the starting element to the end element, and red arrows show the cause of it.|
|Hinders Disturbs||Black arrows show elimination from the starting element to the end element, and red arrows show to disturb it.|
Figure 5: Example of Function Information Analysis Type
3.4. Provision of Limited Small Number of Operators
The objective of the airport management operations is to conduct arrival and departure of aircraft as close as possible to the schedule while handling information. When this is drawn as a Problem Formulator practically with minimum elements, it can be summarized as the basic form shown in Figure 6.
Figure 6: Function Information Analysis Type
Problem Formulator of the Airport Management Division Operations (1)
Then if any phenomenon where some information is not delivered, insufficient or missing the timing, the drawing will be modified as Figure 7.
Figure 7: Function Information Analysis Type
Problem Formulator of the Airport Management Division Operations (2)
The operator necessary for generating ideas for solution from these two Problem Formulators is �Consider transitioning to the next generation of the system that will provide ( ) in a more effective way� which has a quite high level of applicability. Then, if the following eight elements are available in addition to this Operator, the basic viewpoint for solution can be obtained by utilizing 40 Inventive Principles and Separation Principles.
Figure 8: Basic Operators Selected for Problem Solving
in the Airport Management Division
1) Find an alternative way to obtain [the] (A), that provides or enhances [the] (B).
2) Find an alternative way to obtain [the] (B), that does not required [the] (A).
3) Find an alternative way to obtain [the] (C), that does not cause [the] (D).
4) The useful element (C) should exist, and should not exist in order to avoid [the] (D).
5) Find a way to eliminate, reduce or prevent [the] (D), under condition of [the] (C).
6) Find a way to eliminate, reduce or prevent [the] (E).
7) Find a way to obtain [the] (F). This way should not be influenced by [the] (E).
8) Find a way to resolve the contradiction: [the] (G) should exist to obtain [the] (H), and should not exist to avoid [the] (I).
In the course of the program development, attempt was made to solve several problems using this Operator. As a result, for example, in the problem structure of Figure 9, the solution that information will be provided in a proper timing by obtaining information currently delivered by the function Z from other operational process could be obtained.
Figure 9: Problem Formulator of a Certain Problem
3.5. Basic Concept of AFD as a Technique for Recurrence Prevention
The basic concept of AFD consists of the technique for approaching what kind of method is there if the problem will be generated intentionally and the idea that the resources, which can be the cause of the problem, definitely exist near there. Boris Zlotin et al. explained this using the theory that if there are three elements -- oxygen, fuel and temperature --, fire can be caused.
Accordingly, arranging resources to realize the phenomenon, rather than searching the cause of the problem randomly, makes cause analysis easier. This can be applied disregarding the problem is in the technical or non-technical field. As the characteristic of the airport management division, frequency of appearing of human factors as resources just increases.
Effectiveness of AFD is to improve easiness of thinking as well as expand the thinking range by converting the negative thinking process of searching the cause to the positive thinking process of establishing the mechanism and discovering the resources.
The characteristic of this thinking development is to make it possible to easily list elements, which hinder operational functions. If such elements can be identified, it is possible to establish the situation where stable functioning can be realized by eliminating or separating any resource, which causes the element, from the operational process.
For example, a delay of aircraft can easily be caused by [Delay of delivery of number of passengers to be on board]. In other words, in order to realize the delay of aircraft, the method to realize this [Delay of delivery of number of passengers to be on board] without fail should be considered. This method has several resources (not the idea of main cause and second cause, but the countermeasure for the entire system can be considered by reviewing the all the resources). Then the most effective method is [Cause a trouble in the timing of communication]. As the method to eliminate this, [Communication staff will be changed from the current gate operator to the lamp manager] can be determined.
Then, by considering the human errors against this event, [Cause a trouble in the timing of communication], human errors generated here can be identified.
3.6. Whole Picture of the Training Program
The human error prevention training of the airport management division in Company A was designed as the one consisting of two phases.
The first phase is the learning phase of basic thinking (solution thinking of pursuing Ideality based on the TRIZ concept) for errors and problem solving mainly human factors.
The second phase is to learn the details of analysis and solution thinking to prevent recurrence of errors generated.
Figure 10 shows the element structure. In the drawing, TRIZ was applied at elements (3), (4), (5) and (6).
Figure 10: Program Whole Picture - Learning Elements
In the course of developing this training program, by adding TRIZ to recurrence prevention of human errors, it was confirmed that the human error factors could be positioned in the entire system and it was effective for prevention of recurrence as system. However, there is an aspect that in modification of Inventive Principles of TRIZ, complete objectivity could not be maintained. This point should be continuously improved through the actual implementation of training and verification of the results.
Furthermore, even if TRIZ was applied, analysis of human errors was still ambiguous. This point also should be further improved through accumulation of analysis.
Development of this program is supported by the program development project members of Airline A as well as Sadao Saito, senior researcher, Chizuru Takami, researcher, and Masataka Ohta, sales staff of SANNO.
*1 The applicable aircraft are commercial jets manufactured
in the Western countries with maximum take-off weight of 15 t or more.
(back to article)
*2 Revenue flight loss and compensations for passengers are
(back to article)
*3 Professor at Bunkyo University. Searching Social Factors
of Accidents, p. 29, 1992 [in Japanese].
(back to article)
(back to top)