Ready for takeoff: Transforming service parts performance for an aircraft manufacturer

The aircraft engine manufacturing industry can be tough, especially if you don't deliver replacement service parts to clients on time and flights are delayed or canceled. When that happens, margins erode. This manufacturer was regularly paying penalties to its customers due to unfulfilled or outstanding back orders. On the flip side, some parts were clogging up inventory because they weren't moving fast enough. It was clear to the firm that its parts management forecasting wasn't accurate enough – and that led to substantial revenue leakage.

Manual processes and exceptions kept the company's cost per order high. On average, two or three people physically processed about 40% of all orders using Excel. With such outdated inventory tools, out-of-stock situations were common, and inefficiencies ran at 4–5%.

Getting a handle on automation was critical. The manufacturer was capturing 150 million pieces of data for every long-distance flight its engines made – information transmitted in near-real time. It needed a way to incorporate insights from this data into the more than 100 daily engine performance reports it produced for customers. By integrating predictive maintenance plans into its order management and fulfilment processes it would be better able to predict which parts would be necessary where and when.

The company had already tried to resolve some of these issues, but it wasn't getting the results it wanted. A piecemeal approach didn't address customer dissatisfaction or attrition, which produced further losses of 1–3%.

Our first step was to evaluate the strategic problems and possible solutions to make sure suggested changes delivered the right ROI.

We improved demand planning with a safety stock optimization solution to reduce parts forecast error and improve parts case fill rate to ensure that the right parts are available at the right location and time based on inventory demand trends.

Next, we began to digitize supply chain operations. We deployed our intelligent parts forecasting solution to analyze demand patterns, perform statistical forecasting, and manage exceptions from what was budgeted or expected. It determines whether parts and parts segments should be stored on-site or remotely. After this assessment, we created smart processes with built-in machine learning to learn from the results and forecast slow-moving parts.

To improve order fulfilment, we ran design-thinking sessions to select predictive analytics and machine learning solutions to make real-time parts management decisions. Our automated order fulfilment solution streamlines order-to-delivery workflows across systems to enable the efficient fulfilment of customer orders by removing manual processes. Now, artificial intelligence (AI) and mobile apps provide the firm's customers with real-time updates on order queries. The system also has a unique framework that uses AI to identify contract reconciliation issues involving chargebacks, claims, and deductions.

To improve predictive maintenance, we introduced an IoT solution that includes an IoT ROI modeler, a remote monitoring and diagnostic toolkit, warranty analysis, connected field services, and aftermarket services.

Armed with this data, a team of more than 30 engineers in our remote monitoring and diagnostics centers analyzes more than 190,000 operational alarms each year. This real-time data feeds into the parts-planning process and improves asset reliability and performance.

We've integrated, streamlined, and harmonized the service parts planning and replenishment process for this manufacturer and delivered end-to-end benefits across the supply chain:

• The demand planning changes have boosted customer service levels for spares and save more than $20 million each year through better inventory optimization and lower forecast errors
• Our parts analysis for 33 stock keeping units has brought greater visibility into – and greater control over – inventory, leading to efficiencies across reconciliation, reducing obsolescence, and fulfilment
• Supply planning changes delivered $300 million in productivity improvements by centralizing global parts management and fulfilment planning and by removing manual processes There are now 8,000 fewer requests from customer fulfilment teams each year
• The predictive maintenance improvements have reduced engine downtime and enabled better parts planning, saving the firm $250 million each year through fewer delays and cancellations. Maintenance costs have decreased by 40%, and customers have more confidence in engine performance