 |
LIALPA
SlipStream Newsletter, April, 1999 Editor: Capt. R. J. Fitt, IFALPA Director for LIALPA |
| |
LIALPA
SlipStream Newsletter, April, 1999 Editor: Capt. R. J. Fitt, IFALPA Director for LIALPA |
*Click
on any 
to
go back to the top of the page.
Here
are links to the articles - just click, and you're there.
Correspondence
| Culture,
Error and Crew Resource Management
It's a Hell of a life... | Policy,
Trust and Data
Fire the Work... | Focusing CRM on
threat and error
(Special Announcement - LIFAA (formerly LISA) is now on line...)
The Latest
Pay ScheduleAt a meeting with the powers that be on April 20th last, our team from LIALPA were able to learn a bit more about what is going on with reguard to the financial state of our company.
In spite of reports of a positive reception and encouraging noises from our bankers, no financing has yet been approved from them. The matter of alternate sources of finance was to be discussed at a Board meeting held last Tuesday 20th. (At that meeting the matter was assigned to a group of members for their attention, according to an inside source.) The Strategic plan is not yet ready. As I write this, the CEO is in the UK meeting with Speedwing on the matter. Mr. Hadeed felt that the pay schedule should be back to normal in a month.
The matter of T&T representatives on the board is causing problems. Obviously it is not in LIAT's interest for sensitive plans to be discussed with them present. In fact that is all they wanted to discuss at the recent meeting. Clearly that matter might be delayed until the status of the LIAT shares owned by T&T, or their representation, is resolved. Now that the Antiguan Government has expressed serious interest in the acquisition of those shares, there is a sudden reluctance to part with them. I have always felt that it is not in the interests of BWIA and T&T for LIAT to be completely independent, and this is now becoming manifest.
I have also confirmed that since December, LIAT's overdraft has been reduced considerably, from about $13 million to about $3 million. Of course it must be borne in mind that during that period we finally got rid of the Twotters, and had been carrying the usual heavy winter and Easter traffic. Unfortunately, with the Otters gone, we now have less collateral with which to secure additional credit.
So at the end of the day we know a bit more of what is happening, and there does seem to be an effort to make progress towards a return to the contracted pay schedule. The situation is complicated and fluid. The Council will continue to monitor it to ensure that our interests are not compromised.
LIALPA
Unfortunately, only one person was able to get the time off to attend the ALPA sponsored Negotiation Seminar in Washington. Capt. Arthur Senhouse should be back now with his newly acquired expertise from which we hope to benefit. So I guess we can expect a huge increase at our next salary negotiation.
At the last General Meeting, a new Secretary was elected to fill the vacancy left by the resignation of the former holder of the office. Also counted were the votes for Councilmember. The new Secretary is Capt. R. J. Fitt, ( P. O. Box W 787, Antigua; Tel.: 268 461 2580; e-mail: fittr@candw.ag) while there was a tie for the Councilmember position between Capts. Delisle and Sam. That tie will be resolved by a runoff election, the ballots for which will be circulated shortly.
You will notice that the new bid roster, while offering some rotation of Off days, still leaves much to be desired. In fact LIALPA had been in consultation with Ops. Management about a new approach, and had prepared a roster which met or exceeded all the stated criteria, only to be rejected at the last moment for not providing even more reserve capacity than was asked for. This was extremely disappointing after a great deal of work had been put into producing a rostering plan that would have given us a lot more flexibility.
That's all for now, more scoops next month.
Robert L. Helmreich, John A. Wilhelm, James R. Klinect, & Ashleigh C. Merritt, The University of Texas at Austin (Continued from March Slipstream, with interruptions added.)Threat and Error in Flight Operations
Errors have been extensively studied in the laboratory, in training programs, and in post-mortem analyses of crew behavior in accidents and incidents. Similarly, systematic evaluations of the nature of external threats to safety are most frequently investigated in depth when they are associated with adverse events.
There is a dearth of systematic empirical data on the kinds, frequency, management, and resolution of threats and errors in normal flight operations. If safety efforts are to be optimally effective, such information is essential. In an effort to fill this gap, our research group has started a new program to examine threat and error in line audits of normal operations (described in a following section).
The first formal study consisted of observations on 102 international flights at a major airline. To the best of our knowledge, these are the first systematic data linking threat, errors and error management in normal line operations that have been collected under non-jeopardy conditions.
Of the flights observed, 73% experienced one or more off-normal conditions (such as adverse weather, mechanical malfunctions, inadequate ground support, fatigue, language problems with ATC, etc.), with an average of 1.5 per flight (the setting of the study was international flights with challenging environmental conditions). While many of the situations experienced were not serious in themselves, they did increase the level of risk and the probability of error.
When a large number of off-normal events are associated with a particular flight, demands on the crew are greatly increased. On 8% of the flights, four or more off-normal events were recorded. We have found in earlier audits that conditions of high complexity with off-normal conditions may either stimulate crews to superior performance or lead to performance breakdowns (Hines, 1998).
We operationally define error as action or inaction on the part of the crew that leads to a deviation from expectations. Violations of formal requirements such as regulations, SOPs, and policies are also included in this definition. While we recognize the distinction made by Reason and Hudson between errors and violations, we have labeled violations intentional non-compliance errors because we realize that the intent in violations is usually to shortcut what is seen as an unnecessary procedure or regulation or to use a more effective strategy.
In developing a model of crew-based error, we found that the usual taxonomies and classifications of error did not fit our data well. This led us to develop a revised taxonomy of error that we feel may be of value for both research and operational evaluation.
We use a five-way classification of error. Procedural errors include slips, lapses, and mistakes where the crew intended to follow procedures but made an error in execution. Specific procedural errors observed include incorrect entries in the flight management computer and unintentionally skipping items on checklists.
Communications errors include not only mistakes within the cockpit, but also in interactions with air traffic control, such as incorrect readback.
Proficiency errors are reflected in events where one or more crewmembers lack the knowledge to perform a needed action such as a flight management computer procedure or lack necessary stick and rudder skill.As discussed earlier, violations such as skipped checklists, failure to observe sterile cockpit, or not executing a required go-around at decision height are called Intentional non-compliance errors and consist of cases where crews choose to shortcut or ignore procedures.
The fifth category consists of Operational decision errors that reflect situations not covered by SOPs where crews make a decision that increases the level of risk on the flight. These often reflect deviations from policy in cases where there are no formal procedures. Examples include both crew members focusing their attention on reprogramming the flight management on final approach, which is discouraged by the company's automation philosophy.
We also discovered that errors must be classified as detected or undetected in recognition of the fact that many errors may not lead to adverse consequences but, nevertheless, increase the risk of the flight. Undetected errors, by their nature, do not have a crew response.
As for detected errors, crews can respond by ignoring them (in the case of non-compliance), by trapping them, or by taking action that exacerbates the situation - either by leading to another error or to an undesired state, e.g. the aircraft being in an improper vertical or lateral navigation, speed, or fuel condition.
After an error occurs and the crew responds, there is an outcome that can be classified into one of four categories.
An Undesired State is a condition that deviates from normal flight. It includes incorrect navigation, fuel state, unstable approach, long landing, etc.
An outcome is Inconsequential when an error is discovered and trapped without leading to an undesired state. Undetected or ignored errors can also be inconsequential when they have no adverse effects (luck?).
Additional Error refers to an outcome where the initial error leads to a subsequent one, either through no response or an exacerbating response on the part of the crew. For example, failure to run a landing checklist may lead to failure to lower landing gear. After entering an undesired state, the condition can be managed by a crew response that corrects (mitigates) the error or in a manner that exacerbates the severity by leading to another error or to an accident or incident. If one error causes another error at any point, we can start again at the top of the model and the situation represents the classic "error chain".
An average of 1.8 errors were recorded per flight. The distribution of errors, however, is not symmetrical across flight segments. There were no observed errors on 27% of the flights observed, while 12% of the flights had five or more errors.
The highest percentage of errors, 49%, occurred during the approach and landing phase of flight. Boeing's compilation of worldwide jet accidents between 1959 and 1997 comes up with 55% occurring during this phase of flight. The British Civil Aviation (1998) global accident database shows 70% of accidents in the approach and landing phase, but it also includes non-jet and air taxi operations. It is noteworthy that the incidence of error in this phase of flight is associated with the incidence of accidents. In any event, the data validate the importance of proactive steps to reduce risk and error in this phase of flight.
Early CRM advocates fell into the trap of thinking and asserting that it would be a universal panacea for the problem of human error. This did not happen. In today's more restricted, but realistic, model, CRM is seen as a tool that can be used to build a safety culture in the framework of the three cultures that influence flight operations.
This was an "Actual Question" given on a University of Washington chemistry midterm...
"Is Hell exothermic (gives off heat) or endothermic (absorbs heat)? Support your answer with proof."
Most of the students wrote proofs of their beliefs using Boyle's Law (gas cools off when it expands and heats up when it is compressed) or some variant.
One student, however, wrote the following:
First, we need to know how the mass of Hell is changing in time. So, we need to know the rate that souls are moving into Hell and the rate they are leaving. I think that we can safely assume that once a soul gets to Hell, it will not leave. Therefore, no souls are leaving.
As for how many souls are entering Hell, let's look at the different
religions that exist in the world today. Some of these religions state that if you are not a member of their religion, you will go to Hell.Since there are more than one of these religions and since people do not belong to more than one religion, we can project that all people and all souls go to Hell. With birth and death rates as they are, we can expect the number of souls in Hell to increase exponentially.
Now, we look at the rate of change of the volume in Hell because Boyle's Law states that in order for temperature and the pressure in Hell to stay the same, the volume of Hell has to expand as souls are added.
This gives two possibilities:
1. If Hell is expanding at a slower rate than the rate at which souls enter Hell, then the temperature and pressure in Hell will increase until All Hell breaks loose.
2. Of course, if Hell is expanding at a rate faster than the increase of souls in Hell, then the temperature and pressure will drop until Hell freezes over.
So which is it?
If we accept the postulate given to me by Ms. Therese Banyan during my Freshman year, "That it will be a cold night in Hell before I sleep with you," and take into account the fact that I still have not succeeded in that area, then (2) cannot be true, and so Hell is exothermic.
This student got the only A.
An essential for effective CRM is a credible organizational policy that recognizes the inevitability of human error and elucidates a credible commitment to error management. This policy must be built on trust and a non-punitive stance toward error. Rather than seeking to blame and punish those who err, management needs to understand the roots of error in the organization and to develop an array of defenses against future recurrences.
We are in no way advocating that organizations tolerate the intentional violation of their rules or those of the regulatory agency. No organization can expect to survive if it allows its employees to disregard procedures and safety standards.
To specify needed actions and to determine if safety efforts are effective, organizations must have current and accurate data on the state of their operations and the nature and number threats and errors in their operation. To obtain complete and accurate data requires a high level of trust on the part of employees.
They must be willing to share their mistakes without fear of reprisal. Their trust must also include the belief that management will act on safety issues when they are uncovered. If this trust is established, organizations can obtain meaningful data and use them both to guide the development of appropriate training and as a yardstick for assessing trends in performance and error.
Sources of Data on Organizational Performance and Error
Since the accident rate in commercial aviation is extremely low, surrogate measures must be used as safety and organizational effectiveness indicators (Helmreich, Chidester, Foushee, Gregorich, & Wilhelm, 1990).One indicator is pilot performance during formal evaluations by company evaluators or the regulator (the FAA in the USA). Although these data demonstrate that those evaluated have the ability to perform their jobs, they do not reveal how they behave when not under surveillance. Having above average intelligence and valuing their jobs, pilots can adhere strictly to rules when being checked and are also in a state of higher vigilance during evaluation. Although they may not be diagnostic of system performance, checks do have great value as for modeling and reinforcing appropriate behaviors.
Another organizational indicator is performance in training, but this is also an imperfect predictor of behavior during line operations since it also measures the ability of the individual or crew to perform appropriately while under surveillance. Because of these limitations, organizations need to develop alternative sources of data that minimize the jeopardy/best behavior problem.
We will describe three other sources of data that organizations can utilize to gain understanding of the efficacy of their safety and training efforts and to plan the most effective use of their resources. Our view of the data necessary to manage error effectively parallels that of Captain Daniel Maurino of the United Nations' International Civil Aviation Organization. Based on his global experience with air transport, Maurino (1998; in press) concludes that the most valuable data on the health of operations come from the monitoring of normal operations.
Line audits
We have collaborated in the conduct of line audits in a number of airlines (one was the source of the error data discussed earlier). It is our belief and that of participating airlines that such data provide a reasonably accurate and comprehensive picture of line operations. The key to success of an audit is the credible assurance to crews that all observations are without jeopardy and that no information on any crew will be revealed to management or regulators.
In practice, we have trained a group of expert observers from the airline (pilots from training, flight standards, the union, etc.) in the use of our Line/LOS Checklist (Wilhelm & Helmreich, 1996). The team of observers then samples flights in all fleets and types of operations, usually for a period of a month.
That a realistic picture of the operation is being captured is shown by the fact that observers frequently see violations of SOPs and regulations. For example, as part of a line audit we observed instances of failure to complete (or even use) checklists. This was particularly prevalent in one fleet of one airline. Neither line checks nor Federal Aviation Administration inspections had suggested that this might be a problem. The line audit database gives clear guidance to management as to what to emphasize in training and also indicates where problems of leadership or poor safety norms may be present.
Analyses of the aggregated, de-identified data from line audits give the industry insights into ubiquitous problems such as the use of flight deck automation, the variability of performance in the system, and standardization of procedures and practices (Helmreich & Merritt, 1998; Helmreich, Hines & Wilhelm, 1996; Hines, 1998).
Confidential surveys
Organizations can augment line audit data with confidential surveys, often using an instrument such as the Flight Management Attitudes Questionnaire (FMAQ: Merritt, Helmreich, Wilhelm, & Sherman, 1996). Surveys provide insights into perceptions of the safety organization and illuminate aspects of teamwork among flight crews and other organizational elements including maintenance, ramp, and cabin. At the most detailed level, survey data also indicate the level of acceptance of fundamental concepts of CRM among line crews. They also show where differences may have developed between operational units of organizations, such as fleets and bases. Data from surveys can be used effectively to guide curriculum development for recurrent training by helping the organization target the most important operational issues.
Incident reporting systems
Incidents provide invaluable information about points of potential vulnerability in the aviation system. Confidential, national, incident reporting systems such as NASA's Aviation Safety Reporting System and BASIS (British Airways Safety Information System) programs are very useful for the overall system. In the USA, the ASAP (Air Safety Action Partnership; FAA, 1997) concept was designed to give organizations more complete data on incidents in their own operations.
ASAP encourages participation by providing crewmembers with protection from regulatory reprisal for many types of incidents and rapid feedback about organizational efforts to prevent their recurrence. Each reported incident is reviewed by a team (including representatives of management, the pilots' union, and the FAA) which develops a plan of action along with feedback to the reporter.
American Airlines has the longest experience with ASAP and is receiving reports at a rate of over 3,500 per year. As long as crews feel safe in submitting information to programs such as ASRS, BASIS, and ASAP, the data can give organizations an invaluable early warning system about potential threats to safety.
Our research group, in co-operation with several U.S. airlines, has initiated a project to develop a new ASAP form to probe more deeply into human factors issues in incidents. The object of this effort is to generate data that can be combined with those from other sources such as audits, surveys, and training records to provide organizations with a more comprehensive view of their operations and better guidelines for operations and training.
We also recognize the value of data collected during normal operations from flight data recorders under programs such as the FAA's Flight Operations Quality Assurance (FOQA). Such data provide critical information on the nature and location of instances where normal flight parameters are exceeded.
A limitation of flight recorder data is that they provide no insight into why events occurred and the human factors issues associated with them. Line audits, confidential surveys, and incident reporting systems can augment FOQA programs and lead to a better understanding of causal factors.
Using data proactively for safety
The data collected in support of safety can be directly utilized in safety and error reduction initiatives. By examining the categories of error observed in their own observations, organizations obtain a valid report card on the effectiveness of their operation that different elements of the organization can use to plan necessary action. For example, a high frequency of operational decision errors may suggest a need for additional SOPs. Conversely, a large number of non-compliance errors may indicate inappropriate or too many and too complex SOPs (see also Reason, 1997 for discussion of SOPs and compliance).
A fireman is at the station house working outside on the fire truck when he notices a little boy next door.
The little boy is in a little red wagon with little ladders hung off the side. He is wearing a fireman's hat and has the wagon tied to a dog.
The fireman says "Hey little boy. What are you doing?"
The little boy says "I'm pretending to be a fireman and this is my fire truck!"
The fireman walks over to take a closer look.
"Little boy, that sure is a nice fire truck!" the fireman says.
"Thanks mister", says the little boy.
The fireman looks a little closer and notices the little boy has tied the dog to the wagon by its testicles.
"Little boy", says the fireman, "I don't want to tell you how to run your fire truck, but if you were to tie that rope around the dog's neck I think you could go faster."
The little boy says, "You're probably right, mister, but then I wouldn't have a siren."
CRM courses have matured from the presentation of general concepts of team interaction and personal styles to become much more technical and operationally relevant training programs.
Encouraging progress has been made toward the seamless integration of CRM and technical training that was identified as a major goal at the second NASA CRM conference in 1986 (Orlady & Foushee, 1987). One of the outcomes of this movement toward defining CRM in terms of specific behaviors has been a trend toward proceduralization of CRM, requiring interpersonal behaviors and communications as part of technical maneuvers.
The positive side of this is clear guidance for crews as to expected behaviors and, concurrently, the ability to assess and reinforce their practice. There are several negative aspects of proceduralization. One is the possible loss of understanding of CRM's broader, safety goals when it becomes a set of required actions appended to technical maneuvers (Helmreich, Merritt, & Wilhelm, in press).
The second, clearly identified by Reason (1997) is that the proliferation of procedures may serve to reduce compliance. As more and more well-intentioned procedures find their way into operations, they may lose impact and significance, almost inviting violations.
Placing CRM in the framework of threat recognition, error avoidance and error management should help maintain awareness of the organization's commitment to safety. The formal review of known risks and off-normal conditions can be made part of a crew's preparation.
This type of review also represents a readily observable behavior that can be assessed and reinforced. One of the major venues for decision making should be the formulation and sharing of error avoidance strategies in response to recognized threats.
Similarly, detection and management behaviors are usually observable and can be evaluated and reinforced. As Tullo and Salmon (1998) have noted, monitoring and assessing these behaviors present a new challenge for instructors and evaluators, especially those dealing with behavior in normal operations.
CRM training should address the limitations of human performance, a problem evidenced by the high level of denial of personal vulnerability which is characteristic of the professional culture of pilots and other demanding professions.
This denial works to the detriment of threat recognition and acceptance of the inevitability of error. There is empirical evidence that these attitudes can be modified by training [see Helmreich & Merritt (1998) for an example of attitude change about the effects of fatigue on performance]. Awareness of human limitations should result in greater reliance on the redundancy and safeguards provided by team instead of individual actions.
This training can best be accomplished by providing understandable information about the psychological and physiological effects of stress, with examples drawn from aviation experience. The narrowing of attentional focus under stress provides a compelling example of the deleterious effects of stress.
Positive examples of using CRM in crises -- for example, the performance of the crew of United Airlines flight 232 after losing aircraft control following the disintegration of an engine, can build acceptance of team concepts (Predmore, 1992). It is also important to define the nature and types of cognitive errors to which all humans are prey (e.g., Reason, 1990). Making these slips and omissions salient to pilots through everyday, operational examples can also foster endorsement of the use of CRM countermeasures against error.
National culture and CRM
Although threat recognition and error management are universally valued, this does not imply that the same CRM training will work as well in Turkey as in Texas. The rationale provided to flight crews for error management and the description of relevant behavioral countermeasures need to be in a context that is congruent with the culture.
For example, assertiveness on the part of junior crewmembers can be accepted as an effective strategy and practiced comfortably in individualistic, low power distance cultures such as the U.S. In contrast, simply advocating the use of assertion by juniors in many high power distance cultures is likely to be seen as a bizarre and unworkable proposal.
On the other hand, assertive behavior could be acceptable if it is seen as a means of protecting the organization (or in-group) and as a means of saving the face of the captain by keeping him from making a consequential error.
However, there is still much to be learned about fitting training strategies to cultures. We see testing the error management approach as a challenge for both researchers and practitioners and an area where cross-cultural collaboration will be essential.
Correspondence | Culture, Error and Crew Resource Management
It's a Hell of a life... | Policy,
Trust and Data
Fire the Work... | Focusing CRM on
threat and error
 LARGE image |
Front Page | Site
Directory | Mission
Statement | Joint Protocol Officers |  CA-NOTAMs | Caribbean AvWx Observations | Members News | Notices |  Discussion Forum
|  Hiring Requirements  |
 Animated image |
| CAPA | JALPA | Bahamas-ALPA | LIALPA | TTALPA | VNV-NA = This site now is optimised for screens at least 800 pixels wide = |
| Temperature converter: Enter a number, and click outside the box. |
This page was updated on 27 March, 2001
All pages on this site Copyright © 1999, 2000 Caribbean-ALPA
and CanDoo Creative Concepts. All Rights Reserved.
Click HERE to contact the WebMaster.