Northrop Grumman is a global aerospace technology company and one of the largest defense contractors in the world. The company manufactures a variety of aircraft and integrated software-defined systems, including flagship programs like the B-21, Global Hawk, James Webb Space Telescope, E-2 Hawkeye, and F-35.
The majority of executives surveyed by Gallup say that within two years, greater data and analytics skills will be required of everyone in operations, finance, accounting, marketing, and sales in their companies. Northrop’s CEO knew he had to adopt analytics across his company. But the job market for data scientists is stacked against firms, and do-it-yourself, prepackaged training fails to create the community of practice that moves company culture. Northrop needed a data-driven transformation solution.
Northrop selected Data Society out of eight proposals to build a data science academy with the modalities, curricula, and expertise that addressed its needs.
First, Northrop needed training that modalities to accommodate its globally distributed workforce, minimizing cost and maximizing five benefits:
- Student-instructor interaction
- Assistance with ongoing projects
- Development of a community of practice
- Sharing use cases and developing networks within the company
- Delivering immediately applicable skills
Second, Northrop needed training with a curriculum relevant to its engineering use cases, software systems, and technology platforms.
Third, in order to enable a company-wide data-driven culture, Northrop needed a trainer with the mastery to teach basic data literacy and visualization simply and accessibly for those new to the discipline, and also the expertise to instruct advanced data analytics skills like text mining and neural networks.
Data Society fulfilled Northrop’s first need via a tri-modal program of in-person instruction, live-streamed remote training, and self-paced online content. The in-person and live-streaming programs ensured that Northrop Grumman received the five benefits its management needed, while the online platform reduced costs through prep work and post-course attestation. To ensure a relevant curriculum, Data Society custom-designed a program focused on the exact skills and technologies that Northrop’s engineers needed.
To enable a company-wide transformation, Northrop leveraged Data Society’s comprehensive suite of courses for staff ranging from entry-level analysts and non-technical managers to advanced data scientists.
- Data science and visualization in Excel
- Introduction to R programming
- Data visualization in R and dashboards in RShiny
- Unsupervised machine learning
- Supervised machine learning
- Text mining
- Neural networks for image and text analysis
Participants received additional support from instructors after class to guarantee that they could implement solutions in their work.
Results & Testimonials
Northrop successfully created a data-driven culture with a tangible impact. 95% of students surveyed wanted additional data science training; 91% said that the training made them feel like Northrop was investing in their professional growth. 45% maintain regular contact with academy colleagues. And the impact…
Automated data integration reduces dwell time by 90%: After taking the academy’s R course, a small team of Northrop engineers created an RShiny tool that automates the integration of multiple F-35 process data streams. The tool improved data fidelity and reduced supply chain dwell time by nearly 90%.
Quality escape analysis reduces component sustainment cost by 30%: Quality escapes on an F-35 component were flagged by program managers as too high. So Northrop analysts teamed with shop floor engineers to collect and assess new QA data. Their analysis revealed a hidden insight: a simple adjustment to the storage and maintenance procedure could reduce the component’s sustainment costs by over 30%.
J-STARS schedule optimization solves $7m problem: Joint STARS is the only platform in the U.S. arsenal that combines accurate wide-area moving target detection with synthetic aperture radar imagery to track ground targets in all weather conditions from standoff distances. But at one point, half the J-STARS fleet was grounded due to preventable corrosion, resulting in a $7.35 million loss. Northrop turned to its data science trained engineers to write a set of genetic algorithms that optimized the J-STARS maintenance schedule, which saved millions.