Where were you on Thursday, April 29 at 10 a.m. ET? You might not readily recall unless you participated in Military Aviation Logistics & Maintenance Symposium (MALMS). Then you would remember participating in the highly-anticipated panel discussion on Demand Planning for Advancing Military Readiness.
The panel, moderated by Steve Trimble, Defense Editor for Aviation Week Network, brought together three thought leaders.
Organizations in military aviation like Defense Logistics Agency (DLA) use demand planning data to learn from the past and better predict future demand. The financial impact is enormous, but the impact on the warfighter is most significant. Together, supporting the warfighter and delivering maximum readiness for weapon systems and military equipment helps protect the protectors.
Service supply chain planning doesn’t get more challenging than military aviation. Furthermore, optimizing a service supply chain for “readiness” is very advanced. It’s much more than simple calculations to reduce inventory or increase turns. Those are antiquated KPIs and status quo. More sophisticated technology rooted in advanced data science is required to ensure high-stakes weapon systems availability and overall military readiness. The military must be ready to respond to disasters and mitigate crises without notice.
General Day kicked off the session with a provocative question, “how many of you are 69 years old or older?” You would have been alive for the birth of the B-52, which turned 69 on April 15. He added, “How many of us will be around when all the B-52’s land in the boneyard?“
General Day continued, “At DLA, we think about how to not buy too much of the wrong thing and buy just enough of the right thing.” Service parts optimization is the “physics of logistics” as General Day aptly described it.
This underscores the duration of the product lifecycle in military aviation. Supporting equipment with a 70-year lifespan presents many challenges.
Offering perspective from my 20 years of expertise to the panel, I introduced three levels of inventory optimization sophistication.
For DLA, using advanced statistical methods like the moving average method, exponential smoothing method, or EMA, work just fine to generate an accurate forecast of demand to ensure ordering just enough service parts. However, there are a minority of parts for which the demand is very sparse or intermittent. These types of demand patterns are not amenable to the standard forecasting methods. And making matters more complicated, these same parts tend to have a very long lead time, very high costs and have a disproportionate impact on readiness. Even though these parts are fewer in number, their sporadic demand requires larger importance for how to deal with them.
Looking at historical demand consumption sets a good foundation for readiness. Best-in-class organizations utilize forward-looking information to increase readiness. For example, knowing the general direction of deployment/fleet sizes, whether increasing or decreasing, can improve forecast accuracy. Also, understanding OP tempo adjustment directions, whether upwards or downwards, can help prevent organizations from overstocking the wrong parts.
The ultimate challenge is stocking the right parts that contribute to maximum readiness. For a challenge this important, relying on anything less than the best is too risky. With servicemen and women risking their lives and well-being to protect and serve, you must deploy the most capable tools. It’s a compelling reason why Servigistics is a universal standard for Boeing, Airbus, Honeywell, Lockheed Martin, Northrup Grumman, US Air Force, US Coast Guard, and many others in the Federal, Aerospace, and Defense vertical. Servigistics is the only solution proven to deliver maximum readiness. When the lives of servicemen and women are impacted, only the best solution will suffice!
Sharing the outcome of advanced long-term forecasting with suppliers delivers the greatest possible readiness since suppliers often have skin in the game, too, when making procurement decisions.
Applying advanced forecasting to separated demand streams is a best practice. We know that multiple services and equipment can utilize the same part. With this knowledge, establishing separate demand streams and applying the most appropriate forecasting method to each stream leads to the highest forecast accuracy and greatest overall readiness.
Artificial Intelligence (AI) and Machine Learning (ML) can discern patterns in historical data, especially with highly variable, lumpy demand. These patterns are not easy for humans to detect and react to. Many of these parts have a familiar profile. They are expensive, with long lead times and sparse demand. We must be aware of other influences, including high variability, part criticality, mission seasonality, and so on.
With such complexity, the service metrics we use, namely fill rate or issue effectiveness, do not alone do justice to assure fleet readiness and military readiness. It’s not about having the right part or missing the right part. The duration of the stock out is the key influence whether the weapon system is ready or not.
If we use issue effectiveness as the metric, then we don’t give any consideration for whether the wait time was two days for the part or two months due to stock out.
My suggestion is that the service metric to which we are doing inventory optimization ought to be driven more by the stock out duration than issue effectiveness, at least for those parts that we know are the drivers of readiness for the military.
Chris Seymour, Vice President, Military Sustainment with Bell Textron, Inc, added his perspective with a background as a warfighter, “we have exquisite technology nowadays that support the B-52 and C-130, and the Huey and Cobra also come to mind. We have to be as smart as we possibly can to make them as ready as we possibly can make them.”
Maximum readiness is challenging and complex. Managing the readiness of a fleet the size of B-52 or B-20 is difficult. “How do you manage the readiness of a fleet the size of a B-52 or B-20 without going bankrupt,” said Seymour.
Technology plays a vital role in maximizing readiness, and collaboration and communication also help to increase readiness. An example of the V-22 Osprey, plagued with a readiness problem in the rotor component, the rotors were not forecasted accurately. A $20,000 rotor component was failing during a time of war. Through communication and collaboration, the warfighters helped create a plan to reduce rotor failure rates and increase readiness.
Seymour concludes, “An environment of collaboration, communication, and translation is important to solve these complex challenges.”
This session was memorable for great content from the panelists with their diverse perspectives around demand planning. It was also a positive step in the right direction away from a pandemic into an enlightened future.
Servigistics innovations in artificial intelligence, machine learning, big data, and IoT will maximize military readiness and operational excellence.