This article will discuss:
The 2019 Digital Manufacturing Conference, hosted by The Welding Institute (TWI) in the United Kingdom, held roundtable talks about capturing data from the shop floor. The discussion highlighted the difficulties that factory owners face when deciding what to do with captured data sets.
The forum on the actionable data sets has become one of the defining sets of Industrie 4.0 problems that stakeholders in the industrial sector face. These challenges can be categorized under three main sections:
- Deciding what to do with captured data
- The cost of aggregating data
- The big data sets factory operations produce
Industrie 4.0 problems: dealing with a deluge of data
Every digital transformation initiative starts with efforts to capture data from shop-floor assets or operations. Over time, most factory owners realize that capturing data is just the first step to optimizing a specific process, and the need to define an Industrie 4.0 strategy to implement is just as important.
The sheer scale of the data sets captured from the average factory floor has proven to be daunting to first-time implementers of data-capturing tools. In some cases, factory owners simply enable their data-capturing solutions such as the industrial internet of things (IIoT) to run in the background for a few months before deciding on what to do with all this data. In the end, many simply overlook the data and focus on the conventional ways that have been applied to run industrial processes for decades.
The deluge of data coming from digital technologies deployed on the shop floor can be made less daunting by creating clear cut, actionable plans or strategies on what to do with captured data. The 8 major Industrie 4.0 business models or concepts provide a solid foundation for harnessing factory-floor data. The most popular applications of data-capture strategies focus on predictive maintenance, data-driven plant performance optimization, and optimizing machine utilization.
For predictive maintenance strategies, historical machine or equipment data plays an important role in understanding the cause of breakdowns or equipment performance to draw up a maintenance schedule. Thus, for predictive maintenance, no amount of data captured from the shop floor can be considered overwhelming because they bring context to every unexpected equipment failure on the factory floor.
Data-driven plant performance optimization strategies rely on diverse groups of data sets which include machine utilization data, throughput data, inventory data, and material-handling data. Capturing the stated data sets from these diverse processes leads to the need for analyzing big data. In this scenario, an experienced data analyst will be required to make sense of the captured data to optimize the interrelated processes required to maximize the efficiency of an entire plant’s operations. Looking for external expert assistance is recommended to analyze data from complex processes.
Shouldering the cost of data aggregation
The cost of storing, managing, and analyzing the data sets that digital technology captures can be estimated from the amount of data collected from industrial facilities. If historical data for months or years is to be captured and stored then flexible storage and computing resources will be required to do the job. According to McKinsey Digital, 45% of IT projects go over their recommended or proposed budgets.
Captured data must be stored and properly analyzed to get the best out of it, and both processes cost money. The average cost of storing one terabyte of data is $3,351 a year and the average industrial factory produces approximately 347.56 TB annually. This additional operational cost to simply store factory data is considered expensive by many industrial enterprises.
The cost of data storage does not take into account the wages of a full or part-time data analyst or the cost of setting up and configuring data-capturing tools. In summary, a major Industrie 4.0 problem is the cost of capturing, storing, managing and analyzing large data sets.
Devising a data-analysis and implementation strategy
According to Cisco, the lack of internal alignment is a major reason why IT and, by extension, IoT implementations fail. Without an implementation strategy in place, analyzed data may be used to optimize individual pieces of equipment without capturing the bigger picture. In industrial facilities, the bigger picture refers to the entire production line and its interrelated processes.
To take advantage of digital transformation technologies and the data-capturing capabilities they enable, a strategy must be devised before entering the market to purchase either an IIoT solution or any digital transformation hardware/software. The devised strategy will inform the solution purchase process and the ability to leverage captured data.