Manufacturers have traditionally been very successful using data to increase efficiency and quality but are finding that lean production and cost cutting are no longer enough to remain competitive. The goal today is to integrate and gain insights from data across their complex global and often fragmented supply chains.

Manufacturers generate and store data from many sources across the supply chain, including process control instruments, supply chain management systems, and systems that monitor the performance of products after they?ve been sold. Being able to access hidden data and integrate all of this data across multiple sources provides valuable insights and competitive advantage. These insights can lead to improvements in design and production, product quality, forecasting, more targeted products and distribution, and identify hidden bottlenecks in the production process.

Why Big Data Use Cases in the Manufacturing Industry?

Before looking at some specific big data use cases in the manufacturing industry, let's address the role use cases play in big data analytics.

Unless you narrow your query to a specific business challenge that can be revealed by patterns or examples, you won?t get much value from big data. Just having vast quantities of data at hand for analysis doesn?t mean you can extract the insight you need. Use cases force you to narrowly define the question.

A big data use case provides a focus for analytics, providing parameters for the types of data that can be of value and determining how to model that data and perform analytics.

These are some example use cases that illustrate how big data is being used in manufacturing, helping to optimize operations, improve quality and reduce costs.

  • Improving Manufacturing Processes
    • Manufacturing process defect tracking
    • Testing and simulation of new manufacturing processes
    • Support for mass-customization of manufacturing
  • Custom Product Design
  • Better Quality Assurance
    • Product quality and defects tracking
    • Supplier, components, and parts defect tracking
  • Managing Supply Chain Risk
    • Supplier performance data to inform contract negotiations
    • Supply planning
    • Output forecasting
    • Increasing energy efficiency

Detailed use Cases

Take measurements of work-in-progress products to find manufacturing defects as early as possible, while also identifying any potential process or design flaws. Since defects are typically the result of many factors, analyzing long histories of assembly line sensor data can find subtle anomalies that signify product flaws. Apache Hadoop stores long histories of sensor data while also enabling high speed, real-time, early-warning analytics that correlate real-time measurements with other disparate data, then compare to quality models.

Minimize Non-Productive Time (NPT) by monitoring equipment or product utilization in a live environment to identify patterns that indicate imminent failure. For revenue-generating operations equipment, downtime results in significant lost revenue as well as costly repairs. MapR Distribution for Hadoop enables ongoing analysis of an entire system and lets businesses predict when failure might occur, so preventive maintenance can avoid the failure. For consumer products, failures or need for replacement will depend highly on usage patterns, and tracking those patterns help manufacturers to alert customers when their products need specific maintenance.

Track the movement of vehicles and products to identify the costs of various transportation and process options. By using Hadoop to analyze large volumes of historical, time-stamped location data, businesses can calculate optimal delivery routes and enable dynamic rerouting to minimize the impact of arbitrary obstacles like traffic, energy prices and weather. Businesses can also leverage the optimal delivery system as a revenue-generating basis for premium/expedited delivery services to consumers.

Once a product is manufactured and shipped, companies may have little information on its performance. In order to be able to predict potential product component failures, companies leverage MapR Distribution for Apache Hadoop to combine reading from advanced sensors, data feeds from consumer devices, and use Apache Mahout and other analytic methods and libraries to predict the time and cause of future failures.

Real-time parts flow monitoring is the next step after just-in-time supply chain optimization. By attaching sensors to all parts in the production process and tracking them in real time, manufacturers can have a real-time view to their production process. Apache Hadoop provides a cost-effective enterprise data hub for collecting sensor readings and enabling both real-time and batch analysis to optimize production quality and yield.

Product configuration planning helps accelerate production by offering fast delivery times for the manufacture of millions of different product configurations. Through advanced pattern analysis, the most popular configurations can be predicted.

Market pricing and planning can help companies maximize profits. For example, an agricultural company can use Hadoop to analyze crop quality, seasonality, demand and other supply factors, and then farmers can be advised when to bring food to market, and how to plan for the next season.

Managing and optimizing supply chain and logistics is key to success for every manufacturing company on this planet. There is tremendous amount of information that is generated from the Planning and Raw Material procurement to Distribution/Warehousing stage of the process. This information is very valuable to organization to simulate for potential breakdowns and delays in the process. Also every RFID data can now be tracked and assessed to know exact location of product that will allow planning for optimum usage of warehouse space, distribution and delivery methods.

Most of the manufacturing companies have been collecting call center data records (CDR) for warranty and customer complaints. With Big data tools and technology, companies can use the Call Records to know immediately customer discontent using voice/speech analysis or text analytics. In addition, this information can be used to correlate with social media and internal reports from quality and customer/product surveys and competition analysis.

With advancement in technology, every engineering device is now embedded with sensors and RFID that can actively transmit vital information about machines ? machine variables (temperature, oil level, humidity?), production rate, waste metrics, life expectancy and breakdown information. Any production downtime is a potential huge loss of revenue to companies due to loss of production output, cost of repairs and waste generated in the process. All the machine logs can now be used by companies to proactively plan for downtime using real-time information preventing waste, preventing major repairs and minimal downtime

Customer sentiment analysis has crossed boundaries using feeds from Tweeter, Facebook, Google, Blogs, Reviews?. Eminem is officially the first person ever to get over 60 million Facebook fans. Start of this year, Coco-Cola had the most facebook fans ? 35M+. The point is fans are generating major brand awareness for companies by talking to their friends and families and indirectly acting as extended sales team for your organization. Customer Sentiment analysis provides potential to reduce the loyalty decay rate, increase sales by providing vital consumer feedback on products including packaging and distribution.