Read an Excerpt
The Closed-Loop Planning System for Weapon System Readiness
By Richard Hillestad Robert Kerchner Louis W. Miller Adam Resnick Hyman L. Shulman
Rand CorporationCopyright © 2006 RAND Corporation
All right reserved.
This report is about a methodology for more effectively planning and budgeting depot-level repairs of weapon system components. For those readers not familiar with it, depot-level repair (as opposed to base-level repair and intermediate-level repair) is the process of repairing those weapon system components that are shipped from the base or intermediate level to the depot because they cannot be repaired at those other locations. They are commonly referred to as the components that are not repairable at this station (NRTS) from bases and other repair facilities. Some parts cannot be fixed and are "condemned," usually along with placing an order to a contractor for a new component. Those that can be repaired are sent to depot shops specializing in classes of repair, such as repair of avionics components. A given shop at a given depot may repair all of the NRTS components of a given class for a particular aircraft type worldwide. When there are too many repairs demanded of a shop, that shop can increase its capacity through overtime, but at some point the repairs become capacity limited and a backlog will grow. The depot repair budget is ultimately allocated to shops to pay for the resources to make the repairs. Once a component is repaired, it is put into the supplysystem, and that system allocates the "serviceable" component back to a base. The "planning" stage defines and allocates budgets to depots and ultimately to shops based on expected repairs. The "execution" stage involves the decisions about which specific components to repair given an existing demand and capacity limits of shops. This may involve some reallocation of funds because of unexpected repair volume. The execution year is the current year and the planning years are the future years.
U.S. Air Force depot-level repair constitutes a sizeable business. Figure 1.1 shows the fiscal year (FY) 2003 U.S. Air Force working capital fund budget and the portion of this budget allocated to repair activity. The Depot Maintenance Activity Group (DMAG) is responsible for accomplishing the repairs, including contracted work. The Supply Management Activity Group (SMAG) acts as the intermediary between customers and the DMAG. It supplies parts to both the DMAG and customers (e.g., Major Commands [MAJCOMs]) and purchases repairs from the DMAG. The total DMAG planned sales were $6.5 billion. Of that, the sales to the SMAG were planned to be $3.1 billion. By way of contrast, the planned SMAG purchases of parts were planned to amount to one-third of the amount for the repair activity. In addition to sales to the SMAG, a major portion of DMAG activity is programmed depot maintenance (PDM), meaning overhauls of major weapon systems. Other, lesser sales are to other services, to other Department of Defense (DoD) activities, and to foreign military sales (FMS). This report is about the non-PDM portion of the DMAG, the depot repair of repairable components of aircraft, missiles, and engines.
In 2001, Brig Gen Robert Mansfield initiated and supervised the U.S. Air Force Spares Campaign in which he directed five teams to identify "disconnects" in their respective topic areas and recommend corresponding remedies. From the vast number of issues raised, three generalizations relating to repair planning for depot-level repairable parts (DLRs) are of particular interest:
1. Decisionmaking is dominated by financial concerns and is often focused on commodities rather than weapon systems. The allocation of repair resources is generally done without a clear understanding of the impacts on weapon systems.
2. Beginning with MAJCOM budgeting and on through Air Force Materiel Command (AFMC) and Air Logistics Center (ALC) planning to the execution of repairs, there are many points of inconsistency across the processes, including differing views of what is worth spending resources on, varying goals and objectives, and alternative approaches to pricing repairs. Repair capacity is not a factor in planning above the ALC level.
3. Planning and execution processes are disconnected, with at best only weak linkages among them. There is a general lack of monitoring and feedback and no attempts to understand how the decisions made at one level are consistent with those at other levels or to understand when processes are deviating significantly from the higher-level plans.
Ideally, each agency with a stake in depot-level repairs should be acting in ways that are consistent with a single overall plan that considers operational goals and logistics constraints, and the plan should be developed with an understanding of the sensitivities of matching operational goals with allocation of funds.
The Goals of a Planning System for Depot-Level Repair
To correct the problems noted above, a depot repair planning system should:
1. Link repair plans to the operational capability they are expected to provide. Operational capability goals should include those for "boiling peace," major conflict contingencies, and peacetime training. Generally, the repairs required to support an ongoing contingency or known deployments are not included in the planning process.
2. Explicitly take into account logistics constraints such as shop capacity. Weapon systems need a balanced supply of spares for their components. If constraints are ignored, not only are funds allocated that cannot be spent, but funds spent to repair other system components may not improve the weapon system's availability.
3. Plan for the uncertainties in demand for repairs as well as the expected values. Although the supply system plans for pipeline uncertainties, this planning does not provide "protection" for the uncertainty of demands on the depot repair shops.
4. Provide decisionmakers with information on the trade-offs between weapon system performance and the budget allocation to weapon systems by relating expenditures for repair to operational factors. This information provides a basis for assessing the operational consequences when cuts and other changes to the repair budget are considered.
5. Where applicable, provide the capability to consider a holistic budget strategy that includes investments in future capacity, e.g., additional test equipment along with funding near-term repair budgets.
6. For the year of execution, determine the best allocation of funds to the sources of repair as well as estimates of the repairs they will be making.
7. Provide a capability to monitor performance of the repair system and the health of the weapon system as compared with expectations of the plan and thereby relate repair execution to planning.
The "Closed-Loop" Depot-Level Repair Planning System
The current planning system for depot-level repairs develops a requirement based on historical repair demands. The funding of the requirement depends on other priorities within the Air Force budget process, and frequently the requirement is not fully funded. As noted in the previous section, the current planning systems and process do not supply the implications of this shortfall in terms of reduced sortie capability or weapon availability. Thus, this "open-loop" system cannot easily show decisionmakers the consequences of their repair budgeting decisions. A "closed-loop" system would allow decisionmakers to choose budget levels and would provide feedback on readiness implications. Or, it would allow decisionmakers to iterate between levels of readiness and required budgets to see what the budget consequences of desired levels of readiness would be. It would also close the loop between execution and planning by allowing evaluation of the current state of execution against the objectives of the plan. This report describes and demonstrates an approach that we believe satisfies these requirements. We call this approach the Closed-Loop Planning System for depot-level repairs.
There are two requirements for DLRs. The first requirement is for parts that fail during service and must be replaced. The second is parts needed to achieve specified readiness positions ("holes" in weapon systems and establishing war reserve stocks) by a future time. We call these requirements, respectively, "keep-up" and "catch-up." In general, unless there is a surplus of stock, or it is desired to reduce operational aircraft availability, the keep-up requirements must be satisfied, and these demands are random variables with associated uncertainties. Catch-up requirements are policy-driven variables, not subject to randomness, because decisionmakers can choose goals for holes in aircraft and war reserve parts. These, along with the initial state of the system (also deterministic), then define the catch-up requirement. Because the number of failures in operational flying (the keep-up quantity) is random, there is a risk that catch-up goals will not be met. Subject to financial and capacity constraints, the methodology we describe below allocates repair resources in an attempt to meet the policy-driven catch-up goals with high confidence. Through a process of iteration between the user and the model, the weapon system performance implications of various budget levels or "cuts" can be determined.
Organization of the Report
Chapter Two presents a brief critique and description of current repair planning and execution processes and suggests how they might benefit from the closed-loop methodology. Chapter Three describes the formulation of the planning problem that is the basis of our methodology, and Chapter Four describes the solution to the optimization problem. Chapter Five provides illustrations of the prototype database and model for depot repair planning and makes some numerical comparisons with the current process. Chapter Six suggests directions for building on this research and implementing an internally consistent planning process. Appendix A provides the mathematical details of the methodology, and Appendix B describes details of the prototype database used for testing.
Chapter TwoCurrent Processes for Planning Depot Repairs and the Need for the Closed-Loop Methodology
This chapter reviews aspects of the Air Force's current processes for planning and carrying out the depot-level repairs of repairable parts. In the context of describing current processes, we also lay a foundation for the Closed-Loop Planning System methodology that is described in the following chapters.
The Sources of Demand for Depot Repair of Repairable Components
Demands for parts arise for several reasons:
Demands from bases are generated when parts are removed from weapon systems in the course of daily operations and sent to the depot for repair and replacement. Parts are in pipelines and not available for needed installation in weapon systems. Pipelines, by definition, include unserviceable parts in transit to a depot and serviceable parts on the way to bases, parts in repair or waiting to be repaired at bases, and parts at a depot or contractor that are undergoing repair or in unserviceable condition and available for induction to a repair shop. These pipelines absorb a significant number of the Air Force's inventory of parts. Parts are needed for safety stock to buffer bases from the uncertainties of repair demands. Component repair demands are generated from the overhauling of aircraft and engines at a depot. Parts damaged beyond repair are condemned, possibly establishing requirements to acquire new replacements. Parts are needed to stock war reserves, both readiness spares packages (RSPs) and war reserve items held centrally (other war reserve material, OWRM). FMS users and other services require repair support.
Offsetting these demands are serviceable parts owned by the Air Force or expected to be acquired during the planning period of concern, future purchases of additional parts for future planning cycles (that would likely have delivery dates several years out), and depot repairs. If sufficient total numbers of parts, serviceable and unserviceable, are in the Air Force inventory, it is generally less expensive and quicker to meet needs by repairing unserviceable parts than acquiring new ones. Hence, the above-listed demands for parts generally lead to repair demands.
Constraints on meeting the repair needs are the total number of repairable components in the inventory (carcasses), repair capacity, and funds to pay for the repairs.
Stages of Planning and Execution
With respect to depot-level repair, the Air Force performs the following five hierarchical stages of planning and execution, starting with the longest planning horizon (38 quarters) down to daily execution, the shortest:
1. AFMC estimates "requirements" for buying new parts and repairing repairable weapon system components.
2. MAJCOMs plan and program funding to purchase serviceable spares from AFMC.
3. AFMC allocates funds for supporting weapon systems to sources of supply (SOSs) that in turn allocate obligation authority (OA) to sources of repair (SORs). Obligation authority is the authority to expend resources to perform repairs prior to "selling" the repaired component back to the SMAG.
4. Based on anticipated quarterly or semiannual repair programs, the ALCs make plans to have available the necessary resources: carcasses, capacity, repair parts, and funds.
5. Rather than simply carry out a preplanned program, managing daily execution of repairs is a control process, which is the underlying purpose of the Execution and Prioritization of Repair Support System (EXPRESS).
The closed-loop methodology we describe shortly should provide useful inputs to stages 1 through 4. Ultimately, it could also support tracking and replanning during the daily execution of repairs stage.
MAJCOM and AFMC Long-Term Planning
One of the issues raised in the U.S. Air Force Spares Campaign is that the first two stages in the list above are disconnected, with wide differences between views on what portions of the total need for depotlevel repairs are to be funded by each organization. Basically, the MAJCOMs and AFMC use totally different planning methodologies. The MAJCOMs plan and program the purchase of repairs needed for replenishment as a result of operations, essentially what we term the "keep-up" need. From the MAJCOM perspective, the extra repairs to fill pipelines and provide war reserve spares do not fall within its scope of fiscal responsibility. The MAJCOM methodology is called "the AFCAIG process," where the acronym stands for Air Force Cost Analysis Improvement Group. In application it amounts to a dollar cost per flying hour to the estimated flying hours-essentially a regression based on previous demands for repair and flying hours.
By contrast, AFMC's planning process encompasses each of the demands for parts enumerated in the first subsection. Computations supporting this planning are implemented in the D200A system, the Secondary Item Requirements Computation System (SIRS). SIRS estimates "buy requirements" and "repair requirements" by quarter, going out 38 quarters. It is run four times a year. In an attempt to reconcile the MAJCOM and AFMC views of spares requirements, the Air Force established the Spares Requirements Review Board (SRRB). It is at this point that the current process falls short in that it provides limited capability to understand the weapon system readiness implications of the negotiated solution. The methodology underlying the Closed-Loop Planning System model could offer a very useful tool for the SRRB process.
A Closer Look at SIRS
We describe some aspects of the SIRS computation because it is the cornerstone of AFMC's planning and because we intend this discussion to set the stage for the need for our closed-loop modeling.
Excerpted from The Closed-Loop Planning System for Weapon System Readiness by Richard Hillestad Robert Kerchner Louis W. Miller Adam Resnick Hyman L. Shulman Copyright © 2006 by RAND Corporation. Excerpted by permission.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.