Boeing

AI driven transformation

Harish Rao

Harish Rao, senior director of analytics

Every day at Boeing, we generate an enormous amount of data

Imagine you have to read 10 pages of handwritten notes, find symptoms, diagnose the problem, correlate the problems and suggest recommendations. This may sound easy enough to a data expert—but it can be time consuming.

Now imagine this same exercise for 500 million pages. The time, effort, accuracy and cost required would be overwhelming.

At Boeing, we have successfully built and trained machine-learning algorithms that can identify patterns in data, and make recommendations accurately within just a few minutes.

Design

An engineer designing even a simple part faces immense complexity looking at related design, choice of suppliers, overall cost and quality, availability of similar parts, potential for 3D-printed parts, safety and past experience of work orders on the shop floor. This complexity and the non-linear combination of work factors could be handled instead through a digital virtual assistant, powered by artificial intelligence (AI), which leverages real-time data and makes recommendations to the engineer. This recommendation engine could complete tasks much quicker with higher first-time quality.

Further, simulating the design-to-build process through augmented reality, reinforced by predictive analytics, could expose a new way of implementing customer-driven design changes by understanding the quality, cost, supplier choices and available inventory implications before building the product.

Supply Chain

Accurate demand forecasting could drive significant improvement across the entire supply chain. Bringing together data across past forecasts, current demand, availability of raw materials, capacity and quality of suppliers, and by leveraging machine learning algorithms to change the forecast in a dynamic fashion, we can positively affect cost, productivity and quality.

As well, we could use data analytics to combine natural language processing in the contract, past performance of suppliers, and demand forecasting to deploy a dynamic recommended pricing model to the procurement specialists directly affecting our bottom line. This learning could also lead to redefining the right contract to ensure a win-win.

Inventory management is another aspect of supply chain that can be optimized by predictive analytics. To optimize inventory, determining the right threshold is the key. Inventory level has to be adjusted dynamically, similar to demand forecasting.

Smoothing out the inventory across factories and warehouses ensures inventory optimization. New data can further enrich the predictive models and improve the accuracy of inventory.

Factory

Many jobs within the factory require specialized training to complete safely. Automation of these difficult, repetitive tasks through the use of a combination of human and machine robotics could improve safety, productivity and quality.

Deep learning models, an advanced AI technique, address non-linear work interactions such as safety conditions, training, tool/parts availability and work priorities.

This is an area of huge opportunity for any company, including Boeing. Complex jobs can be automated to improve productivity, quality and safety while helping to meet delivery schedules.

To automate and leverage AI, data from sensors on machines can be connected with traditional data such as design, inventory and safety records, to optimize tasks. Instead of simply identifying a task to be automated, a deep learning model can analyze all the data, determine patterns and recommend the best task for automation.

Service

Our aircraft remain in service for several years after they are built. Detecting and preventing problems before they occur builds confidence and paves the way for higher customer satisfaction.

By deciphering usage patterns such as flight conditions, location, temperature, altitude, wind speed and direction, we could predict with confidence when a part needs maintenance, repair or replacement.

By collating the data from sensors, we can deliver predictive solutions to help our customers better plan their operations and reduce total maintenance costs over the life of the  aircraft they purchase from us.

Boeing is driving these innovations through data analytics to lead the fourth industrial revolution.

By Harish Rao, senior director of analytics

    The Internet of Things

    4 stages image

    The Fourth industrial Revolution is predicted to be a combination of cyber-physical systems – that is a blending of the physical, digital and biological worlds. The concept was popularized by Klaus Schwab, founder and executive chairman of the World Economic Forum, who published a book a year ago entitled “The Fourth industrial Revolution.”

    The vision shows a future where intelligent, connected machines operate within a manufacturing system working independently based on objectives and understanding of the complete production chain. Unlike previous changes in industry over the last millennium, new technologies in the combination of systems would impact every aspect of life around the world.

    To learn more, visit the World Economic Forum online at www.weforum.org.