Direct Aircraft Maintenance Costs-Miltary and Commercial

There are two basic inputs to any maintenance repair operation—labor and capital—and these inputs may be partially substituted for each other. The optional or least-cost combination of labor and capital selected to perform a given maintenance service depends on both the productivities and costs of these two inputs. If the price of one rises, it should be partially replaced by the other, relatively cheaper input. Likewise, if the productivity of one input decreases, it should be partially replaced by the more productive input.

The budgetary process, by allocating fixed amounts for personnel (labor) and capital, has constrained the substitution of one for the other. But regardless of these constraints, the relative labor/capital price ratio has increased over time. During the period 1960-68, the basic pay and allowances for military personnel increased by 50 percent, while the interest rate on long-term Treasury bonds, the cost of capital, increased by only 25 percent.

During the same period the retention rates in military units continually declined, and it became necessary to replace departing skilled mechanics with unskilled recruits. This means that at the same time the cost of labor was rising, its productivity was declining. Both of these changes move the military further away from the optimal or least-cost method of producing services if labor-intensive methods continue to be followed.

In an earlier article, a call was sounded for a re-evaluation of the Air Force’s fundamental maintenance philosophy based on the discussion above and the results of previous studies.^l As a bench mark for comparison, it was suggested that the maintenance cost structure prevailing in the commercial airlines be investigated. Although commercial fares, and to some degree revenues, are established by the Civil Aeronautics Board, profits can be influenced by the costs and productivities of the inputs selected to provide airline services. As we will show, the result is that commercial direct maintenance costs per flying hour are approximately 40 percent lower than those for equivalent military transports. The market mechanism, which stimulates the airlines to seek cost-effective methods of operation, does not apply to the military.

This is not to suggest that the airlines have found the point of optimal efficiency, but that they are closer to it than the military. Their operations may even be slightly biased toward the capital side, to show a lower return on investment and thus the need for an increase in fares. The actions of labor unions probably also provide an incentive to operate with less labor. The military, however, should have the same incentive, given the higher training costs and lower productivities that go with declining retention rates.

In addition to the reasons cited, an investigation into the comparative maintenance cost structures prevailing in the military and commercial sectors takes on added importance with the recent emphasis on a volunteer force. Within the Military Airlift Command, for example, 42 percent of the assigned manpower, or 44,000 persons, are engaged in the materiel function, 85 percent of them being employed in maintenance and 13 percent in supply. Under the volunteer force concept, the market cost of manning such a large maintenance force Air Force-wide could be beyond the budgetary constraint. This means that a more efficient method of providing maintenance services must be sought.

A comparison of maintenance cost structures in the military and commercial sectors is no easy task. In the first place, the accounting methods used by the two sectors differ significantly. The airlines carry direct line and depot maintenance costs in one account and general support costs (servicing, cleaning, etc.) in another. Air Force base or line maintenance includes general support man-hours but not depot or material costs. When sufficient data in any category were available for this study, they were subjected to statistical analysis; when insufficient data were available, only general averages could be used to make the cost structures comparable. A concerted effort was made, however, to use only conservative estimates tempered by the opinions of experienced personnel.

In the second place, the missions and aircraft of the commercial and military sectors are somewhat different. To minimize this effect, we chose the maintenance operation of the Military Airlift Command for analysis. The Lockheed C-141 and the Boeing C-135, backbone of MAC’S airlift fleet, are comparable in many respects with large commercial aircraft. The comparison was made with one of the major trunk airlines, which provided detailed labor and material cost data for the analysis. Included in the sample were eight years’ data (1960-67) on the Boeing 720, six years’ data (1962-67) on three Douglas DC-8 engine combinations, and four years’ data (1964-67) on the Boeing 727. Because the data are not published, this airline will be designated “Airline X” in our discussion.

Finally, the capital costs available for this study included only the cost flow of spare parts and the material used in repairable spares. It was impossible to break out an amortized cost of maintenance support capital, such as test equipment, for both sectors. One reason for this is that part of the cost of such equipment is included in the purchase price of aircraft. This includes airborne diagnosis systems, which reportedly save numerous troubleshooting man-hours, and provision for more rapid accessibility to malfunctioning components. In this sense, the analysis admittedly is partial, being confined to comparison of only the direct maintenance costs prevailing in the two sectors.

Despite all these difficulties, we feel that an adequate comparison of military and commercial direct cost structures can be made and that this comparison provides considerable insight into the efficiency of the military operation.

The average direct maintenance costs that prevailed through the 1965-67 period for three commercial and two military airlift aircraft are presented in the accompanying table. These cost estimates are based on data provided by the Military Airlift Command, the Air Force Logistics Command, and the Boeing Company and include both base and depot maintenance. Fiscal year 1968 cost estimates for the C-141, C-135 are presented separately, since these costs are not compatible with previous years because of (1) exclusion of Class V modification costs and (2) a change in the method of distributing common cost items. A description of the calculations used to derive these estimates cannot be included in an article of this length, but the interested reader can find a detailed outline of the calculations in USAF Academy Technical Report 69-2.²

A Comparison of the Direct Maintenance Cost Structures in the Military and Commercial Sectors for the Period 1965-67
						FY 1968 (except Class V Modifications)
Cost Structure	720	DC-8	727	C-135	C-141^a	C-135	C-141^a
Man-hours/ flying hour	19.3	19.2	16.4	40.0	36.0	40.0	31.5
Labor Cost/fh^b	$72	$66	$62	$152	$137	$152	$120
Material Cost/fh	$71	$73	$79	$100	$100	$58	$64
Total Cost/fh	$143	$139	$141	$252	$237	$210	$184
Labor/ material ratio	10.9	.9	.8	1.5	1.4	2.6	1.9
^aNo major repairs were performed on the C-141 during this period; therefore, costs are expected to increase in the future. ^bMilitary man-hours were multiplied by the direct airline wage rate prevailing during the time period to obtain equivalent cost structures.

Excluding the FY 1968 estimates, when Class V modification costs were not included, it would appear that the military used approximately twice as many man-hours per flying hour to maintain its fleet as the airline did. This in itself is indicative of a very labor-intensive approach to aircraft maintenance in the Air Force. Even if the FY 1968 estimates are used, the difference is significant.

To put labor costs on an equivalent basis, military man-hours were multiplied by the direct labor wage rates of Airline X during the period. Material costs for both sectors were already in dollar figures. The table shows that, even in the area of material, military expenditures were approximately 30 percent higher than those of Airline X, giving a total direct-cost difference per flying hour of about $l00—a significant cost differential considering that the C-141 has a flying program of 600,000 hours a year.

The labor/material ratios also show a significant difference between the two sectors. As expected, military operations are very labor-intensive when compared to the airlines. But, in addition, the commercial ratios show a very interesting pattern: although the total maintenance costs for all three commercial aircraft are approximately equal, the labor/material ratio declines with newer models. This implies a sensitivity to the rising cost of labor which results in labor-saving maintenance provisions being incorporated in the newer aircraft, a factor that could not be costed in this study. There is also some evidence of this trend in the military sector.

In summary, the estimates in the table indicate that the direct cost of military aircraft maintenance is significantly higher than that of the airlines. The main reason for this is the extensive use of labor in the military. Even if one argues that Class V modification costs should be excluded and the FY 1968 figures used, the cost difference runs between $40 and $50 per flying hour. The 1968 figures, however, imply a much higher labor/material ratio, indicating an even more intensive labor operation when compared to that of the airlines.

Of course, objections can be raised on the basis that the estimates presented in this article were derived from limited data. We can only state that a consistent effort was made to err, if at all, on the conservative side in deriving these estimates. Another argument may be that a valid comparison of military and commercial maintenance operations cannot be made on the basis of differences in operating techniques, missions, and aircraft. Here we tried to minimize these differences by selecting the maintenance operation of the Military Airlift Command to analyze. MAC’S mission is airlift, the same as the airlines; and its aircraft, the C-141 and C-135, are large four-engine jet transports, most closely resembling those of the airlines. In fact, the C-135 is the military equivalent of the Boeing 707.

Still other objections may be sounded, based on the comparative utilization, size, and age of the two fleets. During the period analyzed, MAC averaged from 6 to 7 flying hours per day on its C-135and C-141 aircraft, whereas Airline X averaged from 9 to 10 hours on its equipment. This could certainly be a factor in explaining some of the difference; but given the fact that Air Force maintenance units are manned primarily on the basis of total programmed flying hours, we feel that it is not as important a factor as some might believe.

In any case, this argument can be somewhat offset by results obtained by a number of researchers showing that maintenance requirements are more correlated with the number of flights than with total flying time, primarily because of starting and stopping stresses.³ Airline X’s flight lengths averaged from one to three hours, depending on aircraft type, while the C-141 averaged 7.64 hours per flight.⁴ This means that for a given total flying time—the deflating variable in this study—Airline X’s fleet was subjected to a greater number of flights than the military fleet. This would lead one to believe that commercial maintenance requirements per flying hour would be comparatively higher.

There is also a common belief that economies of scale accrue to the management of larger fleets. Over the period analyzed, Airline X owned approximately 28 720s and 40 DC-8s, and its inventory of 727s was growing from an average of 65 in 1966 to 96 in 1967. During this same time period MAC had approximately 20 C-135s, and its inventory of C-141s grew from 68 at the first of 1966 to 271 at the end of 1967. Economies of scale could therefore account for the higher cost on the smaller C-135 fleet but not on the larger C-141 fleet, which averaged two to three times that of Airline X’s 727 fleet. It cannot even be said that Airline X obtained its economies by cross-maintenance with other airlines during this period because only 2.5 percent of the direct maintenance costs were contracted outside the firm. Airline X essentially acted as an entity with a total fleet size less than that of the Military Airlift Command. Given this fact, any economies of scale should have accrued to MAC. The reason for the significant difference in costs must be sought elsewhere.

It might also be argued that the C-141 aircraft was still within its stage of “infant .mortality” during the period of comparison, and its maintenance requirements should therefore be higher. Certainly there is some justification for this argument. The 727 maintenance costs decreased from $127 to $120 per flying hour during the period 1965 to 1967. This change, however, is not great, and Boeing studies show that this stage is passed within one year for military transport aircraft; in the case of a majority of the C-141s, it would be by 1968.⁵ It should also be noted that no major depot repair was performed on the C-141s during the period analyzed and that these costs are expected to increase in the future. These costs could offset the lower maintenance costs expected after the “infant mortality” stage, which in the past has been verified only with line maintenance data.

The C-135 was certainly past its “break in” stage and was subjected to major repair. Although this fleet was small, its costs may provide a feel for the costs experienced in the “steady state” or random failure stage in the life of military transport aircraft.

In summary, we do recognize certain validities in the opposing arguments we have outlined. These differences make any comparison of this type quite difficult. We do feel, however, that the cost structures are so radically different that any adjustments in the estimates will not significantly alter the major finding of this study: maintenance operations in the military are extensively more labor-intensive and more expensive than in the commercial sector. Although the comparisons made in this study were based on the maintenance operation of the Military Airlift Command only, the results should not be utilized to single out inefficient use of labor in a single Air Force component. The Military Airlift Command operates under a basic labor philosophy that is prevalent throughout the armed services.

There appears to be considerable slack in some portions of the system, which will permit the military to economize on labor. How can this desired end be brought about? A number of ways have been suggested, both by us and by members of research organizations presently engaged in Air Force maintenance analyses. The four recommendations which we will outline are intended not as an exhaustive listing but as a framework for a start in the right direction.

Two aspects stand out in a more capital-intensive approach. One centers around the more extensive use of maintenance support equipment, the other around the use of spares and material to effect repairs. Maintenance support equipment includes flight line, shop, depot, and airborne equipment, which provide for a more expeditious and reliable diagnosis and repair of aircraft failures. The high cost of depot maintenance (approximately 40 percent of the total) may well be the result of antiquated labor-intensive maintenance techniques that have prevailed for a number of years. Modernization of depot maintenance facilities along more capital-intensive lines may provide considerable savings in labor.

Provisions for increased accessibility to failed components and airborne diagnosis systems can conserve on line maintenance man-hours. The Air Force is already moving in this direction with newer aircraft, such as the C-5, which will incorporate the Malfunction Detection, Analysis and Recording (MADAR) diagnosis system. This movement should be encouraged and expedited whenever technically reliable diagnosis equipment can be developed. It should not be confined, however, to large transport aircraft. The newest fighter aircraft, for example, employ expensive and sophisticated computers for target acquisition, tracking, and firing. Majors Albert E. Preyss and Richard E. Willes of the USAF Academy believe that, as a result of their study of fighter aircraft technology, it may also be possible to program these computers to diagnose maintenance systems, thus making them productive on the ground as well as in the air.

The other side of the capital philosophy is a more intensive use of spare parts relative to labor man-hours. The Air Force tends to expend a large number of unskilled man-hours at the aircraft in attempting to determine which component of a system has failed, whereas airline policy is to remove a whole bank of components for diagnosis and repair in the shop, where more skilled technicians and better test equipment can be concentrated. This policy obviously saves a number of troubleshooting man-hours and does not add appreciably to material costs, as evidenced by the material estimates in the preceding section. One possible reason for this is outlined in the next recommendation.

Finally, with a more capital-intensive approach the expected number of ineffective maintenance actions, and thus the total man-hour workload, should decrease. Studies by the Boeing Company, the Strategic Air Command, and others show that anywhere from 20 to 40 percent of the total maintenance actions performed on a number of B-52 aircraft systems are necessitated by improper diagnosis and repair of previous failures.⁶ As productivity decreases with lower retention rates, this phenomenon could increase. The substitution of capital for labor should result in a lower margin of error.

Scheduled maintenance comprises over 50 percent of the total maintenance workload on military aircraft. The need for such a large number of man-hours to be devoted in this area is now being questioned. The experience in Southeast Asia has shown not only that the sortie rates can be increased with little additional expenditure of man-hours but also that aircraft can fly long beyond their scheduled overhaul periods with no adverse effects.

Along with this, the whole concept of planned replacement for many aircraft components has been challenged by Chauncey F. Bell and Milton Kamins of the RAND Corporation.⁷ As an example they cite a test program conducted by United Air Lines covering engine accessories, electronics, hydraulics, and air-conditioning components. Although the scheduled overhaul program called for the removal of 1200 components, the tests proved that planned replacement on only 33 components was necessary. United Air Lines therefore feels that scheduled component overhauls are rarely necessary.

The elimination of a planned replacement policy can save material as well as labor. This may be the reason material costs in the commercial sector are not higher than those in the military sector. If the “burnout” stage for many aircraft components does not exist, or is far beyond the programmed overhaul time, premature removal requires larger inventories than are necessary.

Additional support for this argument can be drawn from the studies of Major Frank Dyke of the Materiel Analysis Section, Strategic Air Command. An analysis of a large number of aircraft systems revealed that system reliability was some 20 percent lower after a maintenance action than it was after a number of successful flights when a steady failure rate was reached. Major Dyke feels that aircraft systems are inherently reliable and that a large percentage of failures are maintenance-induced. A less protective policy could help eliminate this phenomenon.

Most Air Force line maintenance organizations operate at a 50 percent self-sufficiency level. Economically, this calls for an extravagant use of resources, both labor and capital, when compared to the airlines’ major maintenance base philosophy. The possibility of establishing strategically located major maintenance bases and rotating aircraft through these bases should be investigated. Whether this could be accomplished by a renovated depot system or would require a separate major base concept should also be analyzed. Not only would such a system conserve on a portion of the capital and labor that is now duplicated at each line base but it would also permit consolidation and fuller utilization of the limited number of highly skilled technicians now available to the military.

Airline mechanics are required to perform general maintenance on the entire aircraft whereas the Air Force has resorted to a large number of specialists to perform its maintenance. Given the short length of service for most enlisted personnel and the fact that specialists are more quickly trained, this policy appears to be economically feasible, but it does result in a low utilization of military manpower. It has essentially been necessary to overman Air Force maintenance units for peak loads because it is considered cheaper to queue mechanics than aircraft. This policy results in a large amount of unproductive time being expended by various system specialists. Tradeoffs between the costs of aircraft delay and additional personnel have been perfected by Murray A. Geisler and Chauncey F. Bell of the RAND Corporation and should be investigated as a basis for a more efficient manning policy.⁸

Beyond this, serious consideration should also be given to effecting a more generalist type of mechanic concept. Research in this area has already been initiated by John W. Merck, also of RAND. This concept has intuitive appeal because it would permit a reduction in manpower and a more productive utilization of the remaining force.

Implicit in these recommendations is the belief that their adoption would enable the same output to be produced with a smaller labor force, thereby conserving on the high and rising cost of personnel in the defense budget. If in the past military labor was indeed less expensive than civilian labor and the budgetary process did put effective constraints on the use of capital, Air Force policy was not as irrational as it would at first appear. But times are changing. Military labor is now expensive, and it will grow more expensive in the future. A change in philosophy toward its use is needed.

1. Herman L. Gilster, “Let’s Stop Using Labor as a Free Good,” Air University Review, March-April 1969, pp. 21-32.

2. Herman L. Gilster, An Investigation into the Use of Labor and Capital for Aircraft Maintenance in the Military and Commercial Sectors, USAFA TR 69-2, USAF Academy, 1969.

3. Theodore S. Donaldson and Anders F. Sweetland, The Relationship of Flight Line Maintenance Manhours to Aircraft Flying Hours, RM-5701TR, Santa Monica: The RAND Corporation, August 1968; Herman L. Gilster, A Statistical Analysis of Maintenance Costs on Large Jet Aircraft, unpublished Ph.D. dissertation, Harvard University, February 1968.

4. Support Systems Engineering Division, C-141A Field Experience Summary, D6-57166C-141A, Seattle: The Boeing Company, n.d.

7. Chauncey F. Bell and Milton Kamins, Planned Replacement, P-3052, The RAND Corporation, January 1965.

8. Murray A. Geisler and Chauncey F. Bell, Increased Operational Capability through Logistics Analysis, P-3233, The RAND Corporation, September 1965.

Lieutenant Colonel Herman L. Gilster (USMA, Ph. D., Harvard University) is presently on duty with Hq Seventh Air Force, PACAF, as Chief, Tactical Analysis Division, Directorate of Tactical Analysis. He is on a year’s leave of absence from his position as Tenure Associate Professor of Economics and Management, U.S. Air Force Academy. Since pilot training in 1954, his assignments have been with the 3205th Drone Group, Eglin AFB, Florida; as Air Training Officer, Office of the Commandant of Cadets, USAFA; with 320th Bombardment Wing, March AFB, California, and 96th Aerospace Wing, Dyess AFB, Texas, until assigned to USAFA in 1963.

Major Lloyd Woodman, Jr., (M.B.A., University of Arkansas) is presently serving as an adviser to the South Vietnamese Air Forces, having interrupted his tour on the faculty, Department of Economics and Management, USAFA, which began in June 1967. Previous assignments have been in the personnel career field with the Alaskan Air Command, Air Training Command, and Air Defense Command.

The conclusions and opinions expressed in this document are those of the author cultivated in the freedom of expression, academic environment of Air University. They do not reflect the official position of the U.S. Government, Department of Defense, the United States Air Force or the Air University.

Direct Aircraft Maintenance

Costs—Military and Commercial