7886-444 0014627 AC NO: 120-17A DATE: March 27, 1978
NAT DE TRANSPORT ADVISORY TED STATES OF RAND CIRCULA R#
FAA AERO COMPER
RECEIVED MAY 17 1378
LIBRARY
- MAINTENANCE CONTROL BY RELIABILITY METHODS
DEPARTMENT OF TRANSPORTATION
FEDERAL AVIATION ADMINISTRATION Initiated by: AFS-230
AC NO: 120-17A DATE: 3/27/78
ADVISORY
AND THE ORAN CIRCULAR
DEPARTMENT OF TRANSPORTATION
FEDERAL AVIATION ADMINISTRATION SUBJECI: MAINTENANCE CONTROL BY RELIABILITY METHODS PURPOSE. This circular provides information and guidance material which may be used to design or develop maintenance control programs utilizing reliability control methods.
20 CANCELLATION. Advisory Circular 120-17, dated December 31, 1964, is canceled.
REFERENCES. This circular is appropriate for guidance of certificate holders who operate aircraft in accordance with Federal Aviation Regulations, Parts 121 and 127.
enarese A. FERRARESE Acting Director Flight Standards Service Initiated by: AFS-230
TABLE OF CONTENTS Page No.
CHAPTER 1. INTRODUCTION.
1. Purpose.
2. Authority.
3. Background.
40-12. RESERVED• CHAPTER 2. RELIABILTTY CONTROL FUNDAMENTALS.
13. General.
14. Primary Maintenance Processes.
15. Reliability Control Systems.
16.-24. RESERVED.
CHAPTER 3. PROGRAM ADMINISTRATION. 25
25. General. 25
26. Reliability Program Document. 25
27.-34. RESERVED. 26 CHAPTER 4. PROGRAM APPROVAL. 35 35• Initial Approval. 35
36. Revision Approval. 35 APPENDIX 1. AIRLINE/MANUFACTURER MAINTENANCE PROGRAM PLANNING DOCUMENT - MSG-2 (26 pages) Page iii
CHAPTER 1. INTRODUCTION
1. PURPOSE. This publication provides information on the application of reliability control methods as an integral part of an approved aircraft maintenance program for operators subject to the provisions of Federal Aviation Regulations, Parts 121 or 127.
a. Its primary objective is to provide guidance for development of programs using reliability techniques. It expresses Federal Aviation Administration practice with regard to control programs utilizing these techniques.
b. This circular encompasses the information and criteria contained in its predecessor, AC 120-17, Handbook for Maintenance Control by Reliability Methods. It combines this with information and criteria for the conditionmonitoring process, formerly published in FAA Handbook 8310.4A. The Airline/ Manufacturer Maintenance Program Planning Document - MSG-2, which establishes the criteria for classifying maintenance processes, is included as Appendix 1.
AUTHORITY• The basis for federal regulation of aircraft maintenance is in section 601(a) (3) of the Federal Aviation Act of 1958. For air carriers and commercial operators subject to FAR Parts 121 or 127, this authority is exercised through Federal Aviation Regulations, sections 121.25(b)(6),
121.45(b)(6), and 127 13(b)(7), which require that operations specifications contain time limitations, or standards for determining time limitations, for airframes, engines, propellers, rotors, appliances, and emergency equipment.
3• BACKGROUND• 2• The first generation of formal air carrier maintenance programs was based on the belief that each functional part of a transport aircraft needed periodic disassembly inspection. Time limitations were established for servicing, checks and inspections, and the entire aircraft was periodically disassembled, overhauled, and reassembled in an effort to maintain the This was the origin of the first primary maintenance process discussed in this publication and referred to as "Hard-Time."
b. As the industry grew, matured, and adopted more complex aircraft, literal application of the "Hard-Time" primary maintenance process became obsolete. The industry came to realize that each component and part did not require scheduled overhaul on a fixed time basis, and a second primary maintenance process evolved, referred to as "On-Condition." It is assigned to components on which a determination of continued airworthiness can be made by visual inspection, measurements, tests or other means without disassembly, inspection or overhaul.
Chap 1 Par 1 Page 1
c. Federal Aviation Administration (FAA) control of these programs was accomplished by individual approval of the hard-time or on-condition check periods for the aircraft, engines, and components. The procedures used to adjust these periods were awkward and ourdensome, often inhibiting logical adjustment. lo alleviate this situation, the FAA worked with the airlines to develop more responsive methods of controlling maintenance without sacrificing safety or FAA regulatory responsibility. This method of control was oriented toward mechanical performance rather than to predicting failure wear out points, as was the case in the previous methods. The new method was entitled "reliability control" because its major emphasis was toward maintaining failure rates below a predetermined value; i.e., an acceptable level of reliability
d. The analytical nature of reliability control disclosed and emphasized the existence of components and systems that did not respond to the hardtime or on-condition processes. This led to a third process whereby no services or inspections are scheduled to determine integrity or serviceability. However, the mechanical performance is monitored and analyzed, but limits or mandatory action are not prescribed. This process is entitled "Condition-Monitoring•"
40-12. RESERVED• Chap 1 Page 2 (thru 12) Par 3
CHAPTER 2. RELIABILITY CONTROL FUNDAMENTALS 13• GENERAL•
a. It is intended that characteristics of each operator, i.e., philosophy, consideration of operational and environmental factors, recordkeeping systems, etc., be reilected in his own program. The extent and scope oi each operator's application of reliability control is defined in his reliability program document.
b. There are four general categories of an operator's maintenance program.
(1) Systems/components.
(2) Powerplants/components.
(3) Aircraft/engine checks and inspections.
(4) Structural inspection/overhaul.
All four may be controlled by a composite program, or each may be handled individually. The program can encompass a select group of items from a category without affecting other controls for the remaining items of that category. for example, the basic engine might be maintained by a program that does not include its accessories. The accessories could be on another program or they could be under traditional operations specifications control.
d. Statistical analysis is most effective in its application to systems and components because the occurrence of failures can be readily reduced to meaningful statistics. When alert rates are used in the analysis, graphic charts (or equivalent displays) show areas in need of corrective action.
Conversely, statistical analysis of inspection findings or other abnormalities related to aircraft/engine check and inspection periods requires judgmental analysis. Therefore, programs encompassing aircraft/engine check or inspection intervals might consider numerical indicators, but sampling inspection and discrepancy analysis would be of more benefit.
PRIMARY MAINTENANCE PROCESSES. The three primary maintenance processes utilized by maintenance programs are (1) hard-time, (2) on-condition, and
(3) condition-monitoring.
a. Following are general descriptions of the three maintenance Each program should include specific definitions of the processes it uses and how they are applied. Refer to appendix 1 (MSG-2) and Advisory Circular 121-1A for further definition of maintenance processes.
Chap 2 Par 13 Page 13
(1) Hard-Time (HT). This is a preventive primary maintenance process. It requires that an appliance or part be periodically overhauled in accordance with the carrier's maintenance manual or that it be removed from service.
(2) In-Condition (OC). This is a preventive primary maintenance process. It requires that an appliance or part be periodically inspected or checked against some appropriate physical standard to determine whether it can contime in service. The purpose of the standard is to remove the unit from service before failure during normal operation occurs.
(3) Condition-Monitoring (CM). This is a maintenance process for items that have neither "Hard-Time" nor "On-Condition" maintenance as their primary maintenance process. CM is accomplished by appropriate means available to an operator for finding and solving problem areas. The detalled requirements for the condition-monitoring process are included as appendix 1 to this circular.
b. Complex (multicell) units may be subject to control by two or even all three of the primary processes. The predominant process will determine its classification. For example, the B-747 Modular Package - Stabilizer Control has CM assigned as its primary maintenance process by the MRB report, but a leakage check, which is a conventional O task, is also specified.
c. The basic engine has characteristics that involve all three primary maintenance processes.
(1) Programs that control engine major overhaul intervals consider the engine as a hard-time unit. The overhaul standards are specified by overhaul manuals or other publications that do not identify individual processes as such.
(2) Programs controlling shop maintenance to a "conditional" standard (restoration, etc.,) may classity the engine as on-condition or as condition-monitoring depending on the characteristics of the program. The applicable maintenance processes and their intervals should be designated in (or referenced by) the program document. MSG-2 (ref: appendix 1), discusses the analysis method for assigning maintenance processes. This method was used in the maintenance review board activity for the engines of the widebodied jets. This analytical method, in conjunction with service experience, can be applied to earlier engines.
15. RELTABILIIY CONTROL SYSTEMS. Typical systems used in reliability control are: (1) data collection, (2) data analysis, (3) corrective action,
(4) performance standards, (5) data display and report, (6) maintenance interval adjustment and process change, and (7) program revision. The intent of this section is not to provide a rigid specification but rather to explain the purpose of the systems which the operator can use as a framework for his particular program. The following paragraphs discuss these systems:
Chap 2 Page 14 Par 14
a. Data collection system. This system should include a specific flow of information, identity of data sources, and procedures for transmission of data, including use of forms, computer runs, etc. Responsibilities within the operator's organization must be established for each step of data development and processing. Typical sources of performance information are as follows, however, it is not implied that all of these sources need be included in the program nor does this listing prohibit the use of other sources of information:
(1) Pilot reports.
(2) In-flight engine performance data.
(3) Mechanical interruptions/delays.
(4) Engine shutdowns.
(5) Unscheduled removals.
(6) Confirmed failures.
(7) Functional checks.
(8) Bench checks.
(9) Shop findings.
(10) Sampling inspections.
(11) Inspection writeups.
(12) Service difficulty reports (MRR).
b. Data analysis system. Data analysis is the process of evaluating mechanical performance data to identify characteristics indicating a need for program adjustment, revision of maintenance practices, hardware improvement (modification), etc. The initial step in analysis is the comparison of the data to a standard representing acceptable periormance. The standard may be a running average, tabulations of removal rates for past periods, graphs, charts, or any means of depicting a "norm."
Programs incorporating statistical performance standards alert type programs). Reliability programs developed under Advisory Circular 120-17 and earlier criteria utilize parameters for reliability analysis such as delays per 100 departures for an aircraft system. They incorporate performance standards as described in paragraph (d) of this section. These standards define acceptable performance. When compared with a running graphical or tabular display of current performance they depict trends as well as show out-of-limits conditions. The system performance data is usually reinforced by component removal or confirmed failure data. The conditionmonitoring process can be readily accommodated by this type program.
Chap 2 Par 15 Page 15
(2) Programs using other analysis standards (nonalert type programs). Data that is compiled to assist in the day-to-day operation of the maintenance program may be effectively used as a basis for continuous mechanical performance analysis. Mechanical interruption summaries, flight log review, engine monitoring reports, incident reports, engine and component analysis reports are examples of the types of information suitable for this monitoring method. For this arrangement to be effective, the number and range of inputs must be sufficient to provide a basis for analysis equivalent to the statistical standard programs. The operator's organization must have the capability of summarizing the data to arrive at meaningful conclusions. Also, actuarial analysis should be periodically conducted to ensure that current process classifications are correct.
(3) Summary. The objective of data analysis is to (1) recognize the need for corrective action, (2) establish what corrective action is needed, and (3) determine the effectiveness of that action.
c. Corrective action system. The actions to be taken are a reflection of the analysis and should be positive enough to effectively restore performance to an acceptable level within a reasonable time. The system must include notification to the organizational element responsible for taking the action.
The system should provide periodic feedback until such time as performance has reached an acceptable level. The mechanics of the corrective action system normally encompass methods that have been established for the overall maintenance program such as work forms, special inspection procedures, engineering orders, technical standards, etc. Special provisions should be included for critical failures; i.e., failures in which loss of the functior or secondary effects of the failure impair the airworthiness of the aircraft.
d. Statistical performance standards system. A performance measurement expressed numerically in terms of system or component failures, pilot reports, delays or some other event (bracketed by hours of aircraft operation, number of landings, operating cycles, or other exposure measurement) serves as the basis for the standard. The development of control limits or alert values is usually based on accepted statistical methods such as standard deviation or the poisson distribution. However, some appiications use the average or base line method. The standard should be adjustable with reference to the operator's experience and should reflect seasonal and environmental considerations. The program should include procedures for periodic review of, and either upward or downward adjustment of, the standards as indicated. It should also include monitoring procedures for new aircraft until sufficient operating experience is available for computing performance standards.
e • Data display and report system.
(1) Operators with programs incorporating statistical performance priate data displays, summarizing the previous month's activity. The report Chap 2 Page 16 Par 15
effectiveness of the total maintenance program. It should highlight systems which have exceeded the established performance standards and discuss what action has been taken or planned. The report should explain changes which have been made or are planned in the aircraft maintenance program, including changes in maintenance and inspection intervals and changes from one maintenance process to another. It should discuss continuing over-alert conditions carried forward from previous reports and should report the progress of corrective action programs.
(2) Programs using other analytical standards (nonalert type programs) should consolidate or summarize significant reports used in controlling their program to provide for evaluation of its effectiveness.
These reports may be in the form of computer printouts, summaries, or any intelligible form. A typical program of this type reports the following information:
(a) Mechanical Interruption Summary (MIS).
(b) Mechanical Reliability Reports (MRR).
(c) Listing of all maintenance process and interval assignment. (Master specification)
(d) Weekly update to letter (c) above.
(e) Daily Repetitive Item Listing (by aircraft).
(f) Monthly Component Premature Removal Report (includes removal rate).
(g) Monthly Engine Shutdown and Removal Report.
(h) Quarterly Engine Reliability Analysis Report.
(1) Engine Threshold Adjustment Report.
(J) Worksheets for maintenance process and interval changes (not provided to FAA, but FAA approves process changes).
f. Maintenance interval adjustment and process change system. A major characteristic of reliability control programs is they afford the operator a formal means of adjusting maintenance/inspection/overhaul intervals without prior FAA approval. This does not relieve the operator or FAA of their responsibility for the effects of the program on safety. Procedures for adjusting maintenance intervals should be included in the program.
(1) Maintenance interval adjustments should not interfere with an ongoing corrective action. Special procedures for escalating systems or components whose current performance exceeds control limits should be provided.
Chap 2 Par 15 Page 17
Typical considerations for adjusting hard-time and on-condition intervals are as follows; however, it is not implied that all of these factors be considered for each case:
(a) Sampling.
(b) Actuarial studies.
(c) Unit performance.
(d) Inspector or shop findings.
(e) Pilot reports Methods for adjusting aircraft/engine check intervals should be included if the program controls these intervals and sampling criteria should be specified.
(2) The system should include procedures for initial classification of maintenance processes (HT - OC - CM) and for changes from one process to another. It should also include authority and procedures for changing maintenance specifications and related documents to reflect the interval adjustment or primary process change.
g. Program revision system. The program should include a procedure for revision which is compatible with FAA approvals discussed in chapter 4 of this circular. The procedure should identify organizational elements involved in the revision process and their authority. The program areas requiring formal FAA approval include any changes to the program that involve:
(1) Procedures relating to reliability measurement/performance standards.
(2) Data collection system.
(3) Data analysis methods and application to the total maintenance program.
(4) Process changes:
(a) For programs incorporating statistical performance standards (alert type programs) procedures for transferring components of systems from one primary maintenance process to another.
(b) For programs using other analysis standards (nonalert type programs) changing systems or components from one primary maintenance process to another.
(5) Adding or deleting components/systems.
Chap 2 Page 18 Par 15
(6) Adding or deleting aircraft types.
(7) All procedural and organizational changes concerning administration of the program.
16.-24. RESERVED.
hap • Page 19 (thru 24)