Human Factors Process for Reducing Maintenance Errors

As a result of the 1997 merger with McDonnell Douglas, the Maintenance Error Decision Aid (MEDA) process offered by Boeing is now available to operators of Douglas-designed commercial airplanes and their maintenance organizations. Since its introduction two years ago, a growing number of maintenance organizations for Boeing-designed airplanes have adopted MEDA, which is a tool for investigating the factors that contribute to maintenance errors. MEDA provides a comprehensive approach for conducting thorough and consistent investigations, determining the factors that lead to an error, and making suggested improvements to reduce the likelihood of future errors.

Maintenance errors cost operators of commercial airplanes millions of dollars each year in rework and lost revenue, and present potential safety concerns. For example, aviation industry studies indicate that as many as 20 percent of all in-flight engine shut downs and up to 50 percent of all engine-related flight delays and cancellations can be traced to maintenance error. In response, Boeing developed the MEDA process to help maintenance organizations identify why these errors occur and how to prevent them in the future. Successful implementation of MEDA requires an understanding of the following:

1. The MEDA philosophy.

2. The MEDA process.

3. Management resolve.

4. Implementing MEDA.

5. The benefits of MEDA.

The MEDA Philosophy
Traditional efforts to investigate errors are often aimed at identifying the employee who made the error. The usual result is that the employee is defensive and is subjected to a combination of disciplinary action and recurrent training (which is actually retraining). Because retraining often adds little or no value to what the employee already knows, it may be ineffective in preventing future errors. In addition, by the time the employee is identified, information about the factors that contributed to the error has been lost. Because the factors that contributed to the error remain unchanged, the error is likely to recur, setting what is called the "blame and train" cycle in motion again.

To break this cycle, the maintenance organization's MEDA investigators learn to look for the factors that contributed to the error, rather than the employee who made the error. The MEDA philosophy is based on these principles:

This principle is key to a successful investigation. Traditional "blame and train" investigations assume that errors result from individual carelessness or incompetence. Starting instead from the assumption that even careful employees can make errors, MEDA interviewers can gain the active participation of the technicians closest to the error. When technicians feel that their competence is not in question and that their contributions will not be used in disciplinary actions against them or their fellow employees, they willingly team with investigators to identify the factors that contribute to error and suggest solutions. By following this principle, operators can replace a negative "blame and train" pattern with a positive "blame the process, not the person" practice.

Technicians who perform maintenance tasks on a daily basis are often aware of factors that can contribute to error. These include information that is difficult to understand, such as work cards or maintenance manuals; inadequate lighting; poor communication between work shifts; and airplane design. Technicians may even have their own strategies for addressing these factors. One of the objectives of a MEDA investigation is to discover these successful strategies and share them with the entire maintenance operation.

Active involvement of the technicians closest to the error reflects the MEDA principle that most of the factors that contribute to an error can be managed. Processes can be changed, procedures improved or corrected, facilities enhanced, and best practices shared. Because error most often results from a series of contributing factors, correcting or removing just one or two of these factors can prevent the error from recurring.

The MEDA Process
To help maintenance organizations achieve the dual goals of identifying factors that contribute to existing errors and avoiding future errors, Boeing initially worked with British Airways, Continental Airlines, United Airlines, a maintenance workers' labor union, and the U.S. Federal Aviation Administration. The result was a basic five-step process for operators to follow (see figure 1 for process flow):

An event occurs, such as a gate return or air turn back. It is the responsibility of the maintenance organization to select the error-caused events that will be investigated.

After fixing the problem and returning the airplane to service, the operator makes a decision: Was the event maintenance-related? If yes, the operator performs a MEDA investigation.

Using the MEDA results form, the operator carries out an investigation. The trained investigator uses the form to record general information about the airplane, when the maintenance and the event occurred, the event that began the investigation, the error that caused the event (see "Maintenance Errors" for common examples), the factors contributing to the error, and a list of possible prevention strategies.

The operator reviews, prioritizes, implements, and then tracks prevention strategies (process improvements) in order to avoid or reduce the likelihood of similar errors in the future.

The operator provides feedback to the maintenance workforce so technicians know that changes have been made to the maintenance system as a result of the MEDA process. The operator is responsible for affirming the effectiveness of employees' participation and validating their contribution to the MEDA process by sharing investigation results with them.

Management Resolve
The resolve of management at the maintenance operation is key to successful MEDA implementation. Specifically, after completing a program of MEDA support from Boeing, managers must assume responsibility for the following activities before starting investigations:

1. Appoint a manager in charge of MEDA and assign a focal organization.

2. Decide which events will initiate investigations.

3. Establish a plan for conducting and tracking investigations.

4. Assemble a team to decide which prevention strategies to implement.

5. Inform the maintenance and engineering workforce about MEDA before implementation.

MEDA is a long-term commitment, rather than a quick fix. Operators new to the process are susceptible to "normal workload syndrome." This occurs once the enthusiasm generated by initial training of investigation teams has diminished and the first few investigations have been completed. In addition to the expectation that they will continue to use MEDA, newly trained investigators are expected to maintain their normal responsibilities and workloads. Management at all levels can maintain the ongoing commitment required by providing systematic tracking of MEDA findings and visibility of error and improvement trends.

Implementing MEDA
Many operators have decided to use MEDA initially for investigations of serious, high-visibility events, such as in-flight shut downs and air turn backs. It is easy to track the results of such investigations, and the potential "payback" is very noticeable.

In contrast, according to David Hall, deputy regional manager in the British Civil Aviation Authority (CAA) Safety Regulation Group, a high-visibility event may not present the best opportunity to investigate error. The attention of operators' upper management and regulatory authorities could be intimidating to those involved in the process. In addition, the intensity of a high-level investigation may generate too many possible contributing factors to allow a clear-cut investigation of the event.

Hall has recommended that operators look at the broader potential for improvement by using MEDA to track the cumulative effects of less-visible errors. Providing management visibility of the most frequently occurring errors can, in the long run, produce profound improvements by interrupting the series of contributing factors. According to Dr. Jim Reason, professor of psychology at the University of Manchester, MEDA is "a good example of a measuring tool capable of identifying accident-producing factors before they combine to cause a bad event."

About 60 operators have already implemented some or all of the MEDA process. Participating airlines have reported several benefits, including the following improvements:

"Operator Experience With MEDA Implementation" includes examples of MEDA benefits realized by specific operators and maintenance providers.

The Maintenance Error Decision Aid (MEDA) process offered by Boeing continues to help operators of airplanes identify what causes maintenance errors and how to prevent similar errors in the future. Because MEDA is a tool for investigating the factors that contribute to an error, maintenance organizations can discover exactly what led to an error and remedy those factors. By using MEDA, operators can avoid the rework, lost revenue, and potential safety problems related to events caused by maintenance errors.

Maintenance Errors
After conducting a study of maintenance sites in the United Kingdom in 1992, the British Civil Aviation Authority compiled the following list of the most commonly occurring maintenance errors:

Where to Get Help With MEDA
Boeing representatives have visited more than 90 airplane maintenance organizations around the world to offer assistance with the Maintenance Error Decision Aid (MEDA), which is becoming the industry's generic term for maintenance error investigation processes.

Support for implementing MEDA is available to all operators of Boeing- or Douglas-designed airplanes on a first-come, first-served basis. Support is typically provided over a three-day period and includes:

Interested operators should contact the Maintenance and Ground Operations Systems of the Boeing Customers organization through their Boeing Field Service representative.

Improving Flight Crew Procedural Compliance
Boeing is developing the Procedural Event Analysis Tool (PEAT) to enhance flight operations safety. Similar to the Maintenance Error Decision Aid, PEAT is a structured, cognitively based analytic tool designed to help investigate flight crew procedural errors and develop strategies to prevent future similar errors. PEAT includes both database storage and analysis software. More information on PEAT features and availability will be provided when development is completed in late 1998.

Operator Experience With MEDA Implementation
A 1995 survey in "Airliner" magazine, the predecessor to "AERO," indicated that readers wanted information about operator experience with Boeing products and services. The comments below are from operators in various stages of implementing Maintenance Error Decision Aid (MEDA). - Editor.

In the spirit of looking beyond the person to get to the problem, Cathy Harris, U.S. Federal Aviation Administration liaison for quality assurance, said, "One of the first things I say in an interview is that we're not interested in taking names. I hold up the MEDA form and show them there is no place to record a name on it. The only name that goes on here is my name, so we can tell who conducted the interview." United has also integrated the MEDA philosophy and investigation methods into the human factors training program for all maintenance employees.

Jose Ramos, manager of reliability, said, "MEDA unifies error investigations involving multiple organizations with overlapping responsibilities. The MEDA process allows organizations such as maintenance operations, quality control, engineering, reliability, and human factors to work cooperatively and share their findings, recommendations, and follow-up analysis across organizational lines."

Tom Smith, quality liaison and MEDA administrator for BFGoodrich, reported that technicians involved in an error often rank among a group's top performers. Investigators often report that technicians feel frustrated because they may have been taught to perform a task in a manner that doesn't agree with the maintenance manual procedure. "When they find out that MEDA aims to help correct that kind of problem, rather than discipline them for not following the manual, you can hear a real sigh of relief!" according to Smith.

Bill Ashworth, BFGoodrich vice president of quality and engineering, said, "MEDA is ahead of its time in terms of simplicity and ease of use. It's a great tool. You don't have to be a human factors expert to use MEDA. Its structure and simplicity give you uniform and repeatable results. And, it brings visibility to the whole company at the management level. Every manager knows that his or her manager is looking at the same information."

BFGoodrich has created a computerized database patterned after the MEDA results form. It makes results available to all managers by means of the company's internal network. The addition of the computer database gives management a real-time snapshot of what errors are occurring, why they are occurring, and what actions are being taken.

BFGoodrich used MEDA to investigate a door rigging issue. The investigation determined that an error on a task card contributed to the technician's error, and specific changes were suggested to the training program and maintenance procedures to prevent similar errors. The changes eliminated the problem and a significant amount of rework associated with it. BFGoodrich is expanding the MEDA results form and database to include errors found during operational audits. The objective is to eliminate factors that contribute to errors before they can cause an event serious enough to require upper management or regulatory attention.

According to HAECO's Peter Hayes, manager of quality assurance, "Our internal quality assurance audit process was based on the regulatory requirement to audit the entire maintenance system within a 12-month period. It kept turning up the same results. The MEDA investigation process and findings are helping us aim our audits at the maintenance system where compliance with QA standards is at risk." HAECO has reported an overall reduction in the occurrence of maintenance errors.

Jerry Allen
Human Factors Specialist
Maintenance Human Factors
Boeing Commercial Airplane Group

Bill Rankin, Ph.D
Associate Technical Fellow
Maintenance Human Factors
Boeing Commercial Airplane Group

Bob Sargent
Senior Specialist Engineer
Maintenance Human Factors
Boeing Commercial Airplane Group

return to top | AERO text-only contents | Boeing Home | Commercial
Copyright The Boeing Company. All rights reserved.