AI in Manufacturing: Enterprise Enablement in 3 Use Cases

Use Cases & Projects, Scaling AI Benjamin Libman

The use of AI and advanced analytics has been gaining pace in the manufacturing industry across a variety of core functions, from product design to supply chain management. And while it's important to highlight and understand the host of AI-driven solutions that can be applied to common industry headaches, it's equally important to appreciate the ways in which AI is not just about finding and using the specialty tools that fit niche problems. AI is also about general enterprise enablement — that is, elevating not only data teams but everyone with a stake in data analysis and insights across the organization.

In this blog, we'll run through three use cases that exemplify the principle of general AI enablement at the organization-wide level. From data engineers to line-of-business users, AI enablement means broadening access to data analytics to more users, and improving the power held by those users.

JTI: AI for Enabling a Digital Feedback Loop

More and more firms are adopting AI to facilitate digital feedback loops, including Japan Tobacco International (JTI), a prominent international tobacco product manufacturer. 

JTI faced a significant challenge due to the absence of a dedicated platform for the development of data science and AI projects. In response, the company sought to transform its ecosystem by embracing a collaborative AI platform. The key objectives were to facilitate data science initiatives across functions and markets while simultaneously building a connected AI ecosystem that could be easily leveraged by diverse user groups within the organization.

This use case, facilitated by Dataiku, empowers manufacturing firms to undertake multiple AI-driven applications, including:

Cash Flow Forecasting: Leveraging AI for accurate and efficient forecasting of cash flows, a critical aspect of financial management.

Visual Defects Identification: Utilizing AI algorithms to enhance the accuracy and speed of identifying defects in the manufacturing process through visual analysis.

Van Load Optimization: Applying AI to optimize the loading of vans, streamlining logistics and enhancing operational efficiency.

Supply Chain Optimization: Harnessing AI for optimizing various aspects of the supply chain, improving overall logistics and resource management.

The tangible results derived from the implementation of AI in these use cases include:

FTE Hours Saved in Monthly Forecasting: The application of AI in cash flow forecasting resulted in significant time savings, streamlining the forecasting process and increasing overall efficiency.

Improved Accuracy in Visual Defect Detection: AI-driven visual defect identification enhanced the precision and reliability of defect detection in manufacturing processes.

Workload Reduction Through Automation: Automation facilitated by AI contributed to a reduction in manual workload, allowing teams to focus on more strategic and value-added tasks.

Vincent Delort, FX Risk Management and Capital Markets Manager at JTI, provided a noteworthy testimonial, stating, "With Dataiku and our cloud solution for processing and storing data, we have ideal conditions to firmly anchor Data Science in JTI. The great success and recognition of cash flow forecasting by EACT shows us that we are on the right track." 

Leidos: Software Development Analytics Improvement and Prediction

Companies like Leidos — an engineering technology company involved in the defense, aviation, IT, and biomedical research industries — often don’t have sufficient actionable information and/or analyses to reliably forecast efforts to monitor and assess the production of software development teams. Leidos, in particular, wanted to use analytics to improve agile-based software project execution processes by identifying key drivers of success, and predicting various outcomes.

With the help of Dataiku, companies like Leidos can (and have) prepare and ingest their data into a platform that allows for data exploration and visualization, and furthermore enables them to use machine learning to improve their software development and predict project execution processes. Web-based dashboard deployment makes collaboration a breeze, and allows data teams to save on time, money, and efficiency.

NXP: Making Everyday AI a Reality

NXP, a global leader in secure connectivity solutions in the semiconductor manufacturing sector, recognized the profound impact that modern analytics techniques could have on addressing specific business challenges and fostering a culture of innovation within its already mature and innovative environment. The company aimed to create a pathway for growth and learning through the integration of Everyday AI.

Collaborating with Dataiku and Eindhoven University of Technology, NXP initiated several key initiatives to bring Everyday AI to life:

Citizen Data Scientist (CDS) Upskilling Program

Recognizing the importance of democratizing data science, NXP implemented a CDS upskilling program. This initiative empowered individuals across various departments to acquire data science skills, promoting a broader understanding and application of analytics throughout the organization.

Efficient Transition & Expansion of ML Footprint

NXP leveraged Dataiku's capabilities to streamline the transition and expansion of its machine learning (ML) footprint. This involved integrating ML models into various aspects of their operations, contributing to more informed decision-making and process optimization.

Experiment Tracking & Reproducibility

Dataiku facilitated robust experiment tracking and reproducibility, ensuring that the outcomes of analytics experiments were well-documented and could be reliably replicated. This not only enhanced transparency but also contributed to the continuous improvement of analytical models.

Rationalization of IT Services

NXP rationalized its IT services, consolidating functionalities that were previously distributed across multiple applications. This optimization allowed for more efficient data management and analytics processes.

Convenient Access to Enterprise Data Architecture

The collaboration with Dataiku and the university enabled NXP to establish convenient access to its enterprise data architecture. This encompassed connections to various data repositories, including TeraData, Oracle, and others, streamlining the data integration process.

The use cases enabled by Dataiku's platform included ML-based solutions for forecasting mass market demand and segments, the establishment of a repository for operations research experiments linked to production data servers, and the application of ML techniques in supply chain operations.

The resulting value for NXP was substantial: 

  • Improved Forecasting Accuracy: Enhanced accuracy in forecasting led to significant improvements in cost and revenue management, influencing millions of dollars.

  • Faster Data Analysis: The implementation of Dataiku allowed NXP to analyze complex data relationships more swiftly than ever before, contributing to informed and agile decision-making.

You May Also Like

How to Address Churn With Predictive Analytics

Read More

What Is MLOps?

Read More

4 Do's and Don'ts of Hiring and Upskilling for AI Talent

Read More

Technoslavia: Navigating the Data World in the Age of Generative AI

Read More