The manufacturing industry is on the cusp of another major industrial revolution… which is a good thing, because it needs one. Globalization, fluctuating energy and shipping costs, diminishing resources and complex supply chains are several of the factors having affected the industry over the past couple of decades. To stay competitive in this intense and variable environment, manufacturers in Europe, the U.S. and across the globe are looking to reduce waste and variability in their production processes while dramatically improving their product quality and yield. That is a tall order. The emergence of the Internet of Things (IoT) and Manufacturing Analytics are helping manufacturers get there.
The Industrial Internet of Things
You may have already heard of IoT, as it is one of the most popular business intelligence buzzwords. The term Internet of Things is quickly gaining tremendous buzz across industries. Simply put, in the IoT, sensors and actuators are embedded in physical objects. These sensors can be embedded in anything from roadways to pacemakers. They are then linked through wired and wireless networks, often using the same Internet Protocol (IP) that connects the internet. In other words, everything around us is becoming connected. This isn’t science fiction and it is revolutionizing industries.
In particular, the manufacturing industry is realizing the enormous potential for cyber-physical systems to improve productivity in the production process and the supply chain. The IoT is so revolutionary for the manufacturing industry it has its own subset: Industrial Internet of Things (IIoT). In the IIot, connections exist mainly to produce physical goods for the marketplace as well as to maintain the physical assets of production. These technologies have the ability to drive what Klaus Schwab, German engineer, economist and founder and executive chairman of the World Economic Forum, has coined as the fourth industrial revolution; following the steam engine, conveyor belt and first phase of IT and automation technology.
Manufacturing and Big Data Analytics
Once implemented, IIoT networks churn out substantial volumes of data. After data collection comes the most important part of the IIot, analyzing the data. In manufacturing, the amount of data isn’t the problem. The manufacturing industry has always had a tremendous amount of data. Extremely large datasets are generated through Manufacturing Execution Systems (MES), Enterprise Manufacturing Intelligence software (EMI), industrial systems (i.e., wind turbines and jet engines) and various other systems.
The problem is with what is or isn’t being done with the data. These systems are producing vast troves of process data; however, the data is typically only used for tracking purposes, not as a basis for improving operations. Tracking large amounts of data isn’t big data analytics. In big data analytics, data sources of all sizes are centralized, analyzed and often visualized with the information then used to draw actionable insights. This is much different than tracking inventory through large stagnate spreadsheets that won’t revolutionize the manufacturing industry. It’s the production of actionable insights from data collected through the Industrial Internet of Things that is generating the fourth industrial revolution.
The Manufacturing Business Intelligence Revolution Is Already Here
In Accenture’s 2015 Industrial Internet Insights Report, companies were asked “How important is Big Data analytics relative to other priorities in your company?” 42 percent of manufacturing companies answered Big Data analytics was a top/highest priority. There are an increasing number of success stories coming from the increased investment and focus on business intelligence in manufacturing.
Smart Manufacturing Success Stories
1) General Electric: During one of their Durathon battery plants, over ten thousands of sensors were measuring various data points live. From temperature to humidity, from air pressure to machine operating data, all the data collected in that IoT system allowed GE to monitor production and adjust processes in real time. It also enables to track battery performance back to specific batches of powder and at every step along the process.
2) Harley-Davidson: The Harley-Davidson motorcycle plant in York, Pa., keeps a constant record of the tiniest details of production, such as the speed of fans in the painting booth. When the software detects variables such as fan speed, temperature and humidity are drifting away from prescribed settings, it automatically adjusts the machinery. Harley- Davidson strives to complete a motorcycle every 86 seconds. They use their IoT software to find bottlenecks keeping them from reaching this goal. Recently, Harley managers used their manufacturing analytics to determine that the installation of the rear fender was taking too long. They changed a factory configuration so those fenders would flow directly to the assembly line rather than having to be placed on carts and moved across an aisle.
Manufacturing of All Sizes and Business Intelligence
Manufacturing companies come in all sizes. You don’t have to be a global behemoth such as General Electric to find value in business intelligence software. In fact, to survive in an industry dominated by large manufacturers, small- and medium-sized manufacturers especially need business intelligence. These manufacturers shouldn’t be discouraged by IoT stories of robots, sensors and motorcycles being built in 86 seconds. There is value in data, large and small; however, the data needs to be analyzed properly.
Modern Software as a Service platforms can provide manufacturers with the power and flexibility they need to analyze and visualize their data in the most professional way. These innovative cloud-based BI software programs can combine data across multiple data sources. Manufacturers can use these platforms to explore data sets, visualize trends and patterns, build professional business dashboards and dig deep into their data. Easy-to-use interfaces with drag and drop features mean manufacturers don’t have to stretch resources and hire data scientist to analyze their data. Employees across an organization can be empowered with data, without needed extensive training. A cloud-based infrastructure makes deployment easy and scalable. It also means large investments in IT infrastructure and expensive IT support are unnecessary. With these tools in hand, small- and medium-sized manufacturers can make data-driven decisions and remain competitive.
To capitalize on manufacturing business intelligence software, manufacturers need to set Key Performance Indicators (KPIs). KPIs are business metrics used to evaluate factors that are crucial to the success of an organization. KPIs should be used for benchmarking, to measure progress and performance. There are various manufacturing Key Performance Indicators that can be monitored with BI software. Here are a few manufacturing KPI examples:
- Line speed by product – Use IIoT software to capture machine-level information. Take note of when and how often your line manufactures certain types of products. Then use the software to track the time and effort required to generate meaningful output for each. Pull this data into your KPI dashboard and analyze often. By tracking this KPI, manufacturers can get a better handle on what mix produces the greatest profit.
- Utilization data – With the right dashboard, manufacturers can gain insight into when their factory produces its greatest output. Manufacturing analytics can also segment this data by days, hours, at what mix and with what employees on the floor. By studying the conditions with the best outcomes, manufacturers can seek to reproduce those outcomes on a regular basis.
- Error rates by product and employee – Avoiding mistakes is every bit as important as optimizing mix and hours on the floor. KPIs dashboards can be used to study error rates and then correlate the results by product and employee.
- Assembly speed by product and employee – Manufacturers should slice and dice production data to better understand what products are easier to produce and then use the data to generate discussions with staff and floor leaders. There may be qualitative information not included in the analyses. Manufacturers should dig deep to determine what is special about those products in particular? Would the factory earn higher profits by writing contracts to increase volume? With the right assembly speed, KPIs manufacturers can make data driven changes to their mix.
- Downtime in proportion to operating time – Unscheduled maintenance increases costs and decreases output… which is exactly the opposite of what manufacturers want. This ratio of downtime to operating time is a direct indicator of asset availability for production.
- Percentage planned vs. emergency maintenance work orders – This ratio metric goes hand in hand with the above downtime in proportion to operating time ratio. It is an indicator of how often scheduled maintenance takes place, versus more disruptive/un-planned maintenance.
- Reportable health and safety incidents – Manufacturers need to ensure a safe work environment. If they don’t, the results can be devastating. The Compliance KPI dashboard should include measuring the number of health and safety incidents that were either actual incidents or near misses recorded as occurring over a period of time.
The Industrial Internet of Things and manufacturing analytics are bringing about the fourth industrial revolution. Manufacturers can ensure they are part of Smart Manufacturing by collecting the appropriate data, tracking the right key performance indicators and investing in the right analytics software.