Predictive, prescriptive analytics

With uncertainty surrounding aircraft manufacturing, maintenance, and repairs (MRO), sticking to a timeline while maintaining efficiency is difficult. Most aircraft require some level of daily maintenance, but with many variables in play, no aircraft perfectly fits a maintenance schedule. Despite plenty of data being collected, it is rarely leveraged properly to reduce unknowns, making day-to-day maintenance and depot work complicated. Solutions that address uncertainty provide more comprehensive insight into operations to optimize production and repair schedules.

Problems

Manufacturing challenges are different than MRO challenges. Maintenance needs of one plane can differ drastically from another identical model. No two planes are exposed to the same conditions or usage and therefore do not need the same support on the same schedule.

An aircraft’s location, for example, directly influences the time between maintenance. Other factors – such as incomplete maintenance logs, unexpected issues, fleet usage, age, and weather – also make it difficult to create accurate maintenance schedules. In many cases, unexpected issues are only evident after starting repairs, causing major delays or strain on expensive personnel.

For depots, aircraft may be scheduled for repairs with a three-week turnaround, but it can take more than a year to procure some parts. And if a specialist is not available, repairs will be delayed. With differing levels of experience needed for each airplane and task, not having the right staff will cause major downtime. These challenges create real need to accurately predict companies’ MRO needs and when those needs will occur.

Current solutions

Existing systems attempt to solve these problems, but they don’t quite succeed. Front-line maintenance solutions track each airplane’s usage by tail number. Based on recent usage data, systems automatically project future usage and estimate when maintenance should happen. In practice, the systems inaccurately extrapolate into the future. Variables such as uneven usage can wreak havoc on these predictions. For example, if a plane is parked for a few days, the system may predict no needed major overhaul for years, but for a plane making cross-country trips in that same timespan, the system may predict major maintenance requirements in a month. This clearly does not properly account for variation in aircraft usage.

These solutions also fall short when it comes to staffing forecasts. Without strong analytics, the only way to lower risk is to hire more people than necessary.

Predictive, prescriptive analytics

Predicting likely outcomes and prescribing responses are central to a strong analytics solution, reducing downtime and inefficiency. Solutions such as TruProcess from Lone Star Analysis provide a more accurate outlook. TruProcess provides a unified view of assets using new or existing data collection systems. Data analysis then generates insightful and transparent predictions and prescriptions.

Lone Star Analysis shows spans of uncertainty more accurately account for unknowns than projected trendlines. Instead of looking at every asset and predicting one future, the system considers the fleet as a whole and considers multiple futures, giving a probability to each. This is followed by recommendations and percentages of certainty. This big picture outlook enables more accurate decisions by eliminating assumptions. By considering uncertainty, parts can be ordered in advance, staffing becomes easier and more efficient, and repairs are quicker.

Companies also need to better understand weather’s effects on MRO. Planes located primarily in a desert need maintenance more often than those located in cooler locations. Most depots already use analytics to harness historical weather data to better predict when something such as thermal paste needs to be reapplied. That thinking should be applied to major overhauls as well.

Lone Star Analysis

Depot maintenance

Having the right technicians and mechanics can go a long way toward increasing efficiency. A delay in refurbishment or overhaul stemming from staffing issues creates a ripple effect. The opposite is also true – if a job is done early but the correct technician isn’t scheduled to work yet, there is no benefit from finishing the previous job quickly. On top of providing inventory, maintenance, and staffing guidance, solutions can find and address efficiency bottlenecks. Systems can recommend machinery and expansion investments by building a simulation (digital twin) that analyzes the best future options for productivity and cost.

Benefits of analytics

Each aerospace organization has its own problems, but there are solutions that can make airplane manufacturing and MRO faster and more efficient; make staffing and inventory management easier; increase production with less downtime; and understand when expansion may or may not be needed. Predictive and prescriptive analytics solutions can be tailored to address each issue while leveraging existing data and technology to make it happen.

Lone Star Analysis