The Data-Driven Manufacturer: How Industry Innovators are Bridging the Gap from Legacy Systems to a Modern Infrastructure

Contributed by

7 min read

The manufacturing industry is very unique in many different ways. It has existed for more than 100 years and continues to evolve, benefiting from every major innovation that we have come across along the way—be it the advent of the internet, miniaturization of equipment, electronics, and sensors, advancements in enterprise-wide applications, RFIDs for improving parts monitoring, by-the-minute tracking of goods using advanced fleet management solutions, or complex supply chain management solutions—to name a few.

If we think about automobiles, we have come a long way already from when Ford started making automobiles in 1903 to how modern cars are made in 2018.

Past vs Present

Early days of car manufacturing (early 1900s) and today's car manufacturing (2018)

Manufacturing leaders today seek operational efficiency and productivity, and they are looking at the Internet of Things (IoT) to bring it. Manufacturers have a lot to think about when transforming their business. Operationally, here are a few major improvements manufacturers seek:

  • Lowering the overall cost of poor quality (COPQ) and improving equipment efficiencies by digitizing factories—in other words, building smart factories.
  • Increasing the mean time between failures (MTBF) in equipment by deploying intelligent solutions that allow for predictive maintenance of equipment.
  • Predicting supply shortages and/or demand spikes to near perfection using historical data and by leveraging advanced machine learning (ML) and artificial intelligence (AI) techniques.

That said, the manufacturing industry, as we all know, is extremely cost-sensitive. While any marginal improvements in efficiency or productivity or any reduction in cost of poor quality (COPQ) is a welcome change, plant managers are constantly faced with questions related to return on investment (ROI) when newer technologies are introduced in manufacturing facilities.

Now, a big part in transformation starts with looking at existing applications, bringing data from silos onto a single platform, using business intelligence dashboarding tools to look for patterns, and finally figuring out correlations using advanced machine learning techniques. One of the most important decisions for solution architects probably is to choose the right data platform that can handle the scale of Industrial IoT data, along with the capability to process the data where it is generated. The sooner the data is processed, the faster action the system can take. In a nutshell, you could either have the data processed in the cloud (public or private) or on the edge near the equipment itself, depending upon the amount of data needed for taking accurate action and the urgency of converting insights into action.

Manufacturing firms looking to modernize their infrastructure are (generally) investing in the following key areas:

  • Augmenting the capabilities of their process engineering software—supervisory control and data acquisition (SCADA), programmable logic controllers (PLC) and manufacturing execution systems (MES)—to automate decision-making within the plant.
  • Building a connected factory solution by IoTizing their equipment and developing predictive maintenance schedules with best-in-class failure analysis models. Essentially, maximizing equipment uptime and, at the same time, reducing redundant regular maintenance costs.
  • Building better insights and predictability into their supply chain that allow them to understand how best to procure the available raw material and maintain their profit margins.

CIOs and software architects require a platform that is easy to get started with, seamless to pilot and deploy, and which pays for itself in the long-term—regardless of the use case(s) considered.

Whether it is halving the on-boarding time for new suppliers for MAG45, enabling Audi to integrate existing applications for vehicle development analysis, or increasing the wafer yield for a wireless chip manufacturer at a much lower cost per terabyte, MapR has become the platform of choice for manufacturers globally. In essence, manufacturing firms need a platform that can, at the least, handle time-series data from sensors, structured data from existing operational applications, and unstructured machine logs.

The MapR Converged Data Platform offers the flexibility to work across any cloud, on-prem, or edge technology. The platform can handle operational data (real-time and historical), enables analytics and machine learning, and easily integrates with both legacy systems and modern applications. In addition, with MapR Direct Access NFS™ allowing the existing applications to read/write directly into the MapR Platform and the ability to bring data from all silos into the MapR-DB, the data platform negates the need for manufacturers to fully transition to cloud. These capabilities are instrumental in letting the Manufacturing Operations Head, the CIO, or equivalent construct an end-to-end operational view and truly convert data into action. Add to that, our Analytics & ML capabilities allow architects to experiment with in-place analytics and deploy machine learning models on target cloud environments without having to copy data in separate locations for advanced analytics. Lastly, MapR Edge solves for the challenges specific to scenarios where it is difficult to bring data from remote assets spread across a large manufacturing plant or sites. With a small footprint and reliable replication capabilities, MapR Edge is ideally suited for space- and bandwidth-constrained manufacturing environments, hence addressing issues related to local data collection very common in manufacturing facilities. Used in combination with a core MapR Enterprise deployment (on-premises or in the cloud), MapR Edge empowers organizations to securely process data locally, quickly aggregate insights on a global basis, and ultimately push intelligence back to the edge for faster and more significant business impact.

MapR Edge for IoT

MapR Edge

The diagram below depicts the overall capabilities of the MapR Converged Data Platform.

MapR CDP

To summarize, with MapR, manufacturers can harness the power of all their data. They can strike a good balance between constant production upkeep, assurance of just-in-time delivery of goods, and high yield management on the one hand, while reducing cost of product quality on the other.

Read this whitepaper to learn more about how the MapR Converged Data Platform is helping several industries achieve real business objectives in their transformation journey.


This blog post was published April 11, 2018.
Categories

50,000+ of the smartest have already joined!

Stay ahead of the bleeding edge...get the best of Big Data in your inbox.


Get our latest posts in your inbox

Subscribe Now