Demand planning in a demanding environment. (Supply Support: Air Force Spares Campaign).

The supply chain process consists of many moving parts that must come together to support customer needs. Only when these moving parts are aligned can the supply chain deliver the right product to the customer at the right time and place. The components of the supply chain process include activities from the strategic planning level to the execution level, such as demand and supply planning at the strategic level and workload and transportation planning at the execution level. The activities that occur at the strategic planning level must be aligned with those that occur at the execution level and vice versa. Thus, the importance of planning in the supply chain is significant for execution. Central to the planning process is determining how customer needs will be met in the short, medium, and long terms.

The Air Force logistics processes embrace components of the supply chain and emphasize the need for quick response to the customer in programs such as Repair on Demand. A solid planning capability must be in place to ensure the success of quick-response processes and other execution-level endeavors. Demand planning provides the necessary solid planning capability by aligning expected resources, parts, people, and capacity to anticipated customer requirements. It is the process of translating the warfighter's logistics support requirements into executable logistics plans and schedules. This emphasis on an accurate and responsive planning process is a prerequisite for a 21st century logistics system based on pipeline velocity, lean inventory levels, and repair on demand to support aerospace expeditionary force operations.

The importance of demand planning and the alignment it provides cannot be overstated. To understand why a properly aligned supply chain is key to the warfighter, consider a day in the life situation in which a shop must repair and turn around a mission capability (MICAP) part quickly. A quick-response process sets the stage at the execution level by providing priority guidelines that support the turnaround of the MICAP. However, a behind-the-scenes planning philosophy and environment must be in place so the MICAP can be eliminated. Without a solid planning environment, the repair of the MICAP would be impaired because of unavailable parts or other resources. To meet the customer requirement, quick reaction would be necessary, and the potential for incurring additional costs in the form of last-minute contracting or overtime would increase. Furthermore, the customer requirement still may not be met despite the quick response. Effects of poor demand planning surface in a number of areas: poor repair planning of ten results in cannibalizations, shipment expediting, and schedule changes to compensate for stock outages and back orders. Negative results also can be seen in terms of inventory management (too much or too little stock), pipeline velocity (poor use of capacity for speedy throughput), total costs, and customer satisfaction.

In an Air Force logistics environment, the process of demand planning can be applied in many situations. By anticipating customer needs, demand planning can help a shop plan for the key end items required during an engine downtime incident, minimizing the duration of the downtime. Similarly, anticipating customer needs and planning for these needs can help a shop more effectively align its component parts and avoid awaiting parts (AWP) situations, thereby providing a faster response time. Another situation in which demand planning applies to Air Force logistics is the alignment of capacity and customer requirements. Through the planning process, a shop takes the anticipated customer requirements and examines how current capacity can best meet these requirements. The exercise of balancing anticipated customer requirements with capacity allows the shop to optimize its use of resources, identify shortages and misaligned capacity periods, and produce a realistic execution plan that can be shared with the customer . In turn, the customer can use this information to adjust internal operations.

While the benefits of demand planning are attainable and realistic for the Air Force, the current state of Air Force logistics planning presents notable limitations. The current process is driven by budget requirements and constraints; therefore, it is not a customer-centric process. In the current state, requirements are gathered by several systems that populate the Secondary Item Requirements System (D200A), and from this point, the requirements are aggregated Air Force-wide through the Planning, Programming, and Budgeting System (PPBS) process. These requirements, now in the form of budget figures, are presented to Office of the Secretary of Defense and Congress for approval. During this 2-year process, execution requirements can and do change; hence, requirements become subject to the time line of the budget process instead of being aligned to the needs of the customer.

Current Air Force logistics, at the execution level, operates in a fragmented environment in which different functions--maintenance, supply, and distribution--do not plan in concert. Attempts, such as repair on demand, have been made to improve such inefficiencies, but these have placed emphasis within the isolation of each functional planning area and not in a single logistics planning process across the whole business enterprise. The planning process is centered on a series of unique functional areas rather than on a cross-functional, integrated process. The need of the warfighter or weapon system becomes secondary to the efficiency goal of individual, organizational planning functions. The effect of this functional emphasis has added to the disconnects in overall effective logistics support to the warfighter and the warfighter's weapon systems; for example, insufficient buffers to handle even minor fluctuations in demands. The lack of cross-functionality is also apparent in the design of Air Force legacy d ata systems developed to support individual functions and planning activities. Many of these systems are unable to incorporate information from other systems and are limited in their ability to create integrated plans that span multiple functions. Personnel must spend substantial time compensating for the limitations of the legacy-system environment via significant manual intervention at the expense of providing support to the warfighter and weapon system. Such a tradeoff is unacceptable in today's Air Force.

From the planning process to the supporting systems, current Air Force logistics planning lacks focus on customer demands. Logistics support to the warfighter and weapon systems, from planning to execution and subsequent measurements, is in need of transformation.

Applying Demand Planning

To realize the necessary logistics process transformations, commercial supply chain management best practices that could benefit Air Force logistics have been identified. One of these best practices is demand planning. A combination of process and technology, demand planning provides a methodology that identifies customer needs, establishes a set of stakeholders accountable for planning actions to meet customer needs, manages customer expectations, and builds and continuously evaluates the customer-centric processes that support those needs. A five-step methodology has been developed for executing demand planning.

Establishing the Stakeholders

Step 1. Front-End Agreement. The front-end agreement's objective is to establish a framework for collaboration across supply chain teams (item managers, supply chain managers, shop leaders) that support the customer's requirements. To achieve its objective, the front-end agreement documents roles and responsibilities for supply chain members involved in logistics support of major end items. The framework for collaboration is the foundation for all other activities in demand planning.

In the Demand Planning Pathfinder for the F101-GE-102 engine that powers the B-1B aircraft, the supply chain members established a front-end agreement, via a charter, to provide structure for collaborative planning efforts and senior-level buy-in. It included areas and trigger points for collaboration, the collaborative process that would be used (including finite periods for achieving specific tasks and the level of collaboration), required skills and knowledge level of the participants, and the procedures for resolving disputes. Customers, operations planners, maintainers, suppliers, and other key members of the F1O1-GE-102 engine logistics chain supported the spirit of the front-end agreement throughout the pathfinder test, which demonstrated better support to the weapon system and the warfighter.

Identifying Customer Needs

Step 2. Mission and Logistics Plan. The mission and logistics plan documents the level of weapon system support, logistics issues, and business parameters that would guide supply chain demand planning efforts.

Step 3. Unconstrained Requirements Definition. The unconstrained requirements definition results in a single forecast of warfighter end-item requirements that is used as the basis for all demand planning activities. This number is based on actual warfighter need and does not include financial, capacity, or other constraining factors. Of key significance is the focus on the anticipated needs of the warfighter--an important paradigm shift from budget constraints as the primary focus in the planning process.

During the initial rollout of the demand planning process, the Core Forecast Team (generally materiel management, production management, and supply management personnel) identifies the full customer demand (reparable requests sent from the field or originating at the depot) the depot repair shop will support. This requirement (estimated quantity needed and approximate timeframe for the need) is used as a baseline for establishing integrated execution plans that will best support this forecast demand.

Step 4. Constrained Functional Plan. The constrained functional plan translates the unconstrained demand forecast into an integrated set of functional plans that addresses constraining factors and develops coordinated plans for optimizing logistics system performance, given the constraints. The constrained functional plans are the product of collaboration set forth in the front-end agreement.

In the Demand Planning Pathfinder, the maintenance shop recognized it had only enough capacity to support 14 low-pressure turbine (LPT) rotor repairs per month. As a result, monthly, unconstrained requirements exceeding this capacity would need to be moved to other months to be repaired (preferably earlier than the customer's need date) to avoid degradation in customer support. Because successful repair is dependent on both capacity and getting the required component parts, the supply plan was altered to better match the revised maintenance execution plan, so maintenance would not experience AWP outages during the repair. The result of these changes was a direct benefit to the customer by reducing the number of not mission capable supply incidents, and the repair shop experienced reductions in the average actual flow times for the LPT rotor.

Evaluating the Effectiveness of the Process

Step 5. Feedback Loop. A feedback loop enables the incorporation of regular measurement and accuracy adjustment into planning activities as a function of real-world execution. It is key to any process for the purposes of monitoring progress and making process improvements. Additionally, feedback highlights the essential nature of demand planning as a continuously iterative process of planning, execution, assessment, and improvement in forecasting and meeting demand.

Testing Demand Planning--OC-ALC Demand Planning Pathfinder

The demand planning process, as described, was tested on a small scale at the Oklahoma City Air Logistics Center (OC-ALC), Tinker AFB, Oklahoma, in the F101 engine environment. This demonstration, known as the Demand Planning Pathfinder, focused on creating a reengineered approach to the development of a consensus workload forecast, the development of integrated functional plans to meet a single forecast, the identification of operational constraints, and the use of data for feedback and performance measurement. Results achieved during this demonstration were impressive: LPT rotor throughput increased significantly, logistics response times were reduced from 89 to 44 days, shop flow times decreased from 59 to 45 days, and there were no production stoppages for spare parts identified as having supportability constraints 30 days out. Additionally, the Demand Planning Pathfinder identified a number of significant challenges that need to be addressed to translate test environment results into practical applicatio n within the broader Air Force logistics chain. These challenges include:

* Information Technology. In many instances, the reengineered approach required extensive manual efforts to bring together data and information from individual data systems not designed for integrated information flow. While marginally adequate for a small-scale pathfinder, large-scale implementation requires appropriate enabling systems and technology to reduce manual labor, monitor performance, and provide rapid decisionmaking capability.

* Change Management. Efforts are necessary to enable the dramatic philosophy shift necessary to implement a customer-centric process that focuses on collaboration, cross-functional integration, and identifying and resolving constraints lead time away. Critical factors include a disciplined and thorough review of policies and practices. Commitment to formal training of all personnel involved with these new business practices is also essential.

Overall, the Demand Planning Pathfinder illustrated that demand planning is a commercial supply chain management best practice with the potential to improve Air Force logistics significantly.

Implementing Demand Planning Under the Spares Campaign

Recognizing the results and significance of the Demand Planning Pathfinder, the Air Force committed to apply demand planning principles at all air logistics centers by approving a demand planning initiative as part of the Spares Campaign. The initiative is an ambitious effort to reap the benefits of demand planning across the Air Force via a three-phase plan.

Phase I. Initial Operating Capability 

The demand planning initial operating capability (IOC) focuses on discrete process and policy changes attainable without significant technology resources. This approach emphasizes the change management elements identified in the Demand Planning Pathfinder so that demand planning is adopted as a process first and foremost and then enabled and expanded through technology.

Whether at initial or full operating capability, the core demand planning process is performed at defined intervals (generally monthly for initial operating capability) and is divided into three stages. Each stage is designed to build a specific part of the overall demand plan that will be used during actual execution. The entire process is iterative so that, as it is performed, the Core Forecast Team (generally materiel management, production management, and supply management personnel) uses experience and the assessment of the previous demand plans to influence the development of current plans.

* The planning stage is designed to identify the unconstrained customer demands that represent all anticipated customer needs (estimated quantity and schedule). It also includes the minimum quantity of assets necessary to provide smooth support (pipeline and levels) to satisfy all customer-forecasted demands. This unconstrained requirement is the foundation for developing all future execution plans, which is the next phase.

* The execution stage is intended to build integrated, functional execution plans that are used by each functional area to prepare for the meeting of actual demand for specific items. Once these integrated plans are developed, each functional area will use them to resolve as many potential execution constraints as possible prior to actual execution (for example, induction of the carcass for repair) of the specific items.

* Under the assessment stage, the Core Forecast Team compares actual past execution against what was planned to identify where experience differed significantly from the plan. When the deviation between the previous results and the plan exceed the acceptable parameters, adjustments to the plan are developed for use in the next planning stage. In the IOC phase, a framework establishes the demand planning process through the adoption of the front-end agreement, creation of a core forecast team, an unconstrained requirement, optimization of an integrated execution plan, and a set of metrics.

The Core Forecast Team is formed to develop collaboratively and update (on a monthly basis) the agreed-upon customer forecast for each selected item that should be repaired at the depot. This agreed-upon requirement is known as the one number forecast because the entire supply chain will use it. Starting with historical data, the Core Forecast Team analyzes these data to identify seasonal trends and then creates a forecast, manually or through an algorithm, that projects customer demands 12 months into the future. As part of the collaborative effort, the Core Forecast Team will discuss the validity of this forecast and either accept it or modify it as necessary. Modification of the forecast is a key commercial best practice that recognizes forecasting as part math and part expert knowledge and judgment. This also gives the team ownership of the forecast, instead of surrendering decisionmaking to the system.

The forecast is then used to develop the unconstrained requirement for parts, shop capacity, manpower, and so forth. Key inputs to the creation of the one-number forecast and the unconstrained requirement are the projected monthly customer requirements and any external requirements (planned operational or maintenance events, pipeline assets, safety-level assets, readiness spares packages). The keep up portion of the unconstrained requirement is generally the customer's minimum projected needs, projected events (time compliance technical orders), and any projected establishment or increases in the levels. The catchup portions of the unconstrained requirement are the holes in the existing pipelines or levels and the immediate fulfillment requirement to achieve a new asset level.

Once the unconstrained requirement has been identified, the team's efforts shift to the development of constrained execution plans. Constraints include, at a minimum, maintenance capacity (average available shop and event-driven capacity) and supplier capability (delivery schedule, available volume). Each functional constraint is applied, and all functional results are then reconciled simultaneously to develop an integrated execution plan that best supports the unconstrained requirement while minimizing the risk of producing excess assets. The integrated execution plan is then used to align production to capacity and piece parts to meet customer requirements and measure the performance of demand planning results.

The demand planning metrics are designed to measure the success of the plans and corresponding pre-execution actions on meeting the warfighter's needs while operating within the availability of existing resources. Favorable outcomes should be seen through increased production, which results in experiencing fewer high-priority back orders (occurrence or duration), reduced shop-flow times (including delay times) much closer to the standard values, and improved use of the shop resources. Overtime, the average AWP occurrences and duration should decrease noticeably, while the forecast accuracy (end item and components) should improve. Additional metrics will focus on the Core Forecast Team's ability to anticipate customer needs (forecast accuracy) and how the projected execution plan deviates from actual production. As the Core Forecast Team gains experience and understanding of the demand planning process and collaboration effort, the variance between the forecast and actual demand, as well as projected executio n plan and actual production, will decrease.

Rollout of demand planning initial operating capability will serve a dual purpose: introduce the demand planning process at the air logistics centers and serve as the platform for expanding demand planning across the supply chain.

Demand Planning Rolled Out

The Air Force has embraced the demand planning concept, and the process was introduced at the Warner Robins ALC (WR-ALC) in March 2002, as an extension to ongoing workload planning activities, and designed to reduce monthly variability in shop inductions.

Key to the identification of the Core Forecast Team members is management-level support of demand planning and workload planning. In meetings with management and working-level personnel, the collaborative nature and customer focus of demand planning was accepted, and the front-end agreement was created to support future activities.

A workshop was conducted in June 2002 to launch the demand planning process. During this workshop, team members collaborated to validate historical data and create a one-number forecast 12 months out, determined capacity during that period, formulated an unconstrained requirement that incorporated anticipated catchup and keep up, and created a realigned maintenance execution plan. Building on this process, the WR-ALC Core Forecast Team has selected 1 low-altitude navigation and targeting infrared for night item and 12 ARC-164 items. Although a nascent process thus far, the workload planning efforts indicate that demand planning will yield significant benefits through back-order reductions and improvements in resource allocation and customer service.

Phase II. Full Operating Capability

The establishment of a demand planning initial operating capability will introduce the air logistics centers to the demand planning process and prepare them to expand the scope of demand planning to additional shops and across the supply chain. Key to this expansion will be introduction and selection of technology to enable the demand planning process. The introduction of technology at this stage will ensure that demand planning is capable of maintaining its emphasis on the process and overall focus on customer requirements. The Core Forecast Team will leverage its combined experience during initial operating capability and apply lessons learned to select the best technology that will transform Air Force logistics to a customer-centric process. This will mark the beginning stage to reach full operating capability.

Demand planning full operating capability will expand the scope of initial operating capability through the selection of additional end items, review of the planning bills of materiel information to support demand planning projected execution plans, and the analysis and selection of information technology tools to support the added complexity and scope of full operating capability across the Air Force Materiel Command.

Phase Ill. Advanced Planning and Scheduling 

To capitalize on the momentum of the logistics transformation Demand Planning Pathfinder at OC-ALC, an APS Pathfinder effort was launched in February 2002.

While the process of demand planning may manifest itself in different forms, either in process or technology such as advanced planning and scheduling, the benefits realized in industry and observed in the Demand Planning Pathfinder cannot be ignored. The fundamental activities of collaboration and proactive planning are constant and provide proactive support to the customer whether or not software technology is being employed. The process of demand planning lays out the framework for collaboration and the information technology, which enables this process and allows an organization to expand the scope of demand planning throughout its supply chain. Any organization whose primary focus is meeting the customer's requirements is compelled to recognize the value of effective planning and execution to meet customer needs. To have multiple organizations work in collaboration for the single goal of supporting the warfighter is the essence of demand planning.

Figure 2

Historical Forecast Data

      Mo 1  Mo 2  Mo 3  Mo 4  Mo 5  Mo 6  Mo 7  Mo 8  Mo 9  Mo 10  Mo 11

Data   21    25    18    22    20    25    27    25    21    20     57
Data   22    12    21    35    17    26    25    18    28    41     59
Data   12    24    23    32    19    20    26    17    24    33     57

      Mo 12

Data   50
Data   45
Data   52

Note: Table made from Line graph

Figure 3

Projected Monthly Customer Requirements

       Unconstrained Rqmt  Rqmt Plan (Keep Up)

Mo 2           15                  15
Mo 3           18                  18
Mo 4           23                  23
Mo 5           17                  17
Mo 6           20                  20
Mo 7           25                  25
Mo 8           17                  17
Mo 9           21                  21
Mo 10          23                  23
Mo 11          57                  57
Mo 12          45                  45

Note: Table made from bar graph

Ms Kaczmarek is project manager, Demand and Repair Workload Forecasting Initiative, Directorate of Supply Chain Integration, Air Force Deputy Chief of Staff Installations and Logistics; Ms LaRue is a logistics officer, Propulstion Production Division, OC-ALC, Tinker AFB, Oklahoma; Mr Anderson is an item manager, General Electric Engines Section, Logistics Propulsion Management, OC-ALC, Tinker AFB; Mr Owen is an industrial technician, Engine Production Branch, Maintenance Directorate, OC-ALC, Tinker AFB; Mr Warren is a program analyst, Materiel System Group/Software Information Tinker Unit, Air Force Material Command! Electronic Systems Center, Tinker AFB; Mr Waite is a logistics officer, Logistics Management Section, Logistics Propulsion Management, OC-ALC, Tinker AFB; and Ms Womack is an inventory management specialist, Maintenance Directorate, OC-ALC, Tinker AFB.