Predictive analytics can provide possible outcomes of a situation. Prescriptive analytics – arguably the most crucial phase in system analysis – supplies optimal outcomes to pursue based on the best solution of each considered result. A simple example is the self-driving car. At each intersection, the vehicle must analyze all possible options, choosing the one to take based on the most desirable end. This model requires analytical units to think for themselves. In the aerospace industry, where original equipment manufacturers (OEMs) and maintenance, repair and operations (MROs) are under tight deadlines to ensure the performance and safety of airplanes, this analysis model could improve efficiency.
In the aviation industry, to combat fluctuating fuel costs and changing flyer demand, many airlines have introduced the concept of rightsizing to their fleet management. This model uses predictive and prescriptive analytics to determine the best plane (regional jet versus a larger aircraft) to be used at any given time, allowing airlines to reach better efficiency standards for fuel usage and ticket offerings. Prescriptive analytics also influences fare determination, with airlines identifying the maximum ticket price a reasonable number of customers will pay at any given time before they are deterred. FLYR, a business-to-business travel company, predicts ticket prices and seat availability using prescriptive analytics to book at the lowest fares for consumers. FLYR’s system allows shoppers to lock in the lowest predicted airfares – even if those happen to be in the future.
MRO efficiency improved
Thanks to the rapid rise of the Internet of Things (IoT), aerospace companies now can collect massive amounts of data on everything from arrival times, fuel usage, and weather impacts, to the performance of miniscule parts within a plane engine. What can airlines and engineers garner from this data?
Until recently, airlines have been using this data for predicative purposes such as identifying the lifetime of a part. With the introduction of prescriptive analytics, computer systems can go beyond predicting possible outcomes to assigning optimal solutions for those outcomes. Predicative maintenance can tell an MRO when a part is most likely to fail. The MRO must then determine, weighing a variety of factors, such as part age, frequency of use or importance to plane performance, whether that part should be repaired or replaced. With the use of prescriptive maintenance, the possible outcomes and solutions to a predicted part failure can be analyzed, and the MRO can be provided with the optimal solution.
According to Airbus, grounding an A380 can cost nearly $1 million a day. Even delays of a few hours are tens of thousands of dollars. In 2015, mass cancellations due to weather at Chicago, Boston, and New York airports cost airlines an estimated $16 million dollars. While weather may not yet be a surmountable impediment, these examples illustrate the massive losses airlines face when their planes are grounded. By implementing prescriptive analytics to the maintenance equation, MROs can quickly gather potential problems and solutions and be ready to attend to incoming planes the moment they hit the ground.
Human-generated solutions to aircraft manufacturing are marked by inefficiency. The human mind cannot grasp all of the information needed to accurately predict and provide solutions to many possible outcomes. Data gathered in aircraft assembly can be analyzed to provide solutions to slowdowns, adjusting quickly to new information or interruptions, such as shipment delays, to provide best solutions.
Obstacles to adoption
While prescriptive analytics have revolutionized many aspects of the aviation industry, adoption for aircraft maintenance has been somewhat stilted. This is primarily due to the interruption the incorporation of prescriptive maintenance models will have on existing MRO activities. Before this model can be included in airplane MRO, several obstacles will need to be addressed.
Regulatory Approval – Aviation businesses must undergo extensive government regulation with major changes to routine maintenance schedules often determined by regulatory bodies.
Software Upgrades – For in-house and independent MROs, prescriptive analytics requires major software upheavals. Existing integration of IoT and the widespread use of handheld tablets by maintenance professionals for diagnosis and repair will ease this transition.
Staff Changes – Skilled professionals will be needed for upkeep of highly technical systems, increasing staff and associated expenses.
Despite these obstacles, International Data Corp. predicts that by 2020, 50% of all data-driven, decision-making software will incorporate prescriptive capabilities.
Aerospace supply chain
Similar to the use of IoT and predicative analytics, the benefits of prescriptive maintenance to MROs has implications for the aerospace supply chain as well. Implementation of this model can assist suppliers in identifying customer issues and providing quick solutions. Suppliers can track the lifetime of particular parts and systems and then predict when MROs will be in need of these parts again – all while taking into consideration factors such as frequency of aircraft use, changes in fleet requirements, and upgrades to existing parts. This reduces the amount of storage need for surplus parts, preventing overstock of parts that may become obsolete. At Kapco Global, an international parts distributor for MROs and OEMs, similar efficiency solutions include real- time inventory figures and reorder dates through its e-commerce solution, Kapco kart. The location of final destinations and warehouse locations are also considered to optimize shipping times.
Vikram Bhatt, COO of Kapco Global explains, “The concept of prescriptive analytics at all levels of operations is an exciting use of the vast amount of data already being collected by the industry. This model allows for continuous improvement in distribution intelligence, making the supply chain a valuable participant in providing efficiency to our MRO partners.”