Data-Driven Challenges For Industrial Manufacturers

Shamanth Shankar
Shamanth Shankar April 17, 2017

Four challenges that block the road to a successful implementation of big data and data analytics at industrial manufacturers.

Industrial manufacturers inhabit a world littered with uneasiness. Global demand for manufactured products is growing at a snail’s pace. Output was expected to increase by just 3.1 percent in 2016 and is estimated to increase by 3.4 percent in 2017, according to the International Monetary Fund.

In a slow-growth environment such as this, productivity gains are paramount. And that could be a boon for industrial manufacturers. Indeed, industrial manufacturers can best serve their customers (and themselves) by designing tools and equipment that improve the efficiency, costs, and performance of factories and other capital projects.

For customers, the desire for efficiency and quality improvements are a given, but companies increasingly want visibility deep into their supply chains: They want connectivity tools that provide insight into production levels, inventory and capacity availability, quality levels, and order status from all their suppliers.

In their study ‘The Biggest Challenges of Data-Driven Manufacturing’, Harvard Business Review [1] analyzed the top four challenges that block the road to a successful implementation of big data and data analytics in the manufacturing industry.

Time-Triggered Versus Event-Triggered Control Systems

ProductionMost of the today’s manufacturing is centered around stable technological systems based on pre-determined levels of demand.

Usually, the production process is triggered by a specific demand schedule which is automatically fed into the ERPs and the manufacturing execution systems (MESs) which pull all the necessary resources from the supply chain. After the production ends, the finished product is packaged and shipped to the final customer through various distribution channels.

The long-term aim of the manufacturing companies is to move on from these time-triggered systems towards more efficient event-triggered control systems where the company produces only when the customer orders the product. The factory will be required to respond to the data signal (the order) only when it occurs, allowing the company to utilize their capital efficiently and configure the systems to collect data and communicate it where and when needed.

Data Sharing, Not Just Data Exchange

IntegrationAnother significant challenge for the manufacturing enterprises is to create a unified data model and tie in together all the independent systems in the manufacturing process (starting with design, engineering production, distribution, and selling). Data needs to be integrated and shared to every business unit in order to minimize the wasted materials and activities.

An excellent example of system integration is how one drill manufacturer has been helping its customers — chemical plants, oil refineries, and other process manufacturers — operate their plants more effectively by leveraging IoT in the form of wireless drill sensors, which can detect potential failures in valves before they lead to a spill or shutdown.

Handling Legacy Systems

Introducing and integrating new technologies within well-established technical ecosystems can bring about new data challenges. These tend to arise not only when migrating information from one system to anther, but also when trying to understand how older systems fit with the modern developments. The lack of a well-defined interface, proper documentation, and different programming languages could be a barrier towards smooth integration within the existing design and manufacturing environment.

Security Challenges

SecurityLastly, all the systems which are interconnected via the Internet are susceptible to data breaches.

This is not yet the case for industrial control systems, which often have unique protocols and limited computing power. Many of these systems were designed and installed at a time when industrial security was not a prime concern. They are connected via gateways that bridge the internet and the controlled device [1]. Still, these gateways need to be able to handle networking and security risks and be strong enough to avoid unauthorized access to data.

Industry 4.0

If connected machines — the primary components of the Internet of Things (IoT) — are to be the backbone of industry in the near future, industrial manufacturers will have to figure out how to manage the data coming from an avalanche of sensors, integrated equipment and platforms, and faster information processing systems.

An MDM solution

An MDM solution serves as a hub for the enterprise data by integrating data from multiple sources (including legacy systems) into a single repository (the single ‘source of truth’) and help manufacturers to manage data related to products, customers, and distribution locations by sharing it across the business. An MDM solution works for both time-triggered and event-triggered systems, integrates easily with legacy systems and enables restricted access to various user groups through data governance rules.

Read about how Top Industrial Producer uses Riversand’s MDMCenter to Achieve Global Repository Functionality Missing from Traditional ERP and Homegrown Systems

Top Industrial Producer uses Riversand’s MDMCenter to Achieve Global Repository Functionality Missing from Traditional ERP and Homegrown Systems

Read 2017 Industrial Manufacturing Trends at pwc.smh.re

Reference:

[1] https://hbr.org/2016/05/the-biggest-challenges-of-data-driven-manufacturing

 

Popular Blogs

Close

 

Show Buttons
Hide Buttons