Industry 4.0, Industrial Internet of Things, and industrial analytics are the next-generation categories currently receiving major attention and investment in our industry. Uses range from predictive maintenance, track and trace supply chain analysis, cybersecurity, and more. As trends in data utilization advance, so do business practices like software-as-a-service, tiered pricing models, software modules, etc. The combination of new technology with the go-to-market strategy of technology providers complicates the entry point for companies deciding on an analytics investment.
Analytics, when thoughtfully evaluated and implemented, can lead to high-impact insights. On the other hand, any automation project can be subject to time and cost overruns, and—given the complexity of connecting systems in order to gain those sought-after insights—analytics projects might be even more susceptible.
Investment in an analytics platform is often more daunting than some vendors would lead you to believe. While many platforms can produce the same graphics and metrics, implementation costs should not be overlooked. Analytics is not just the math and statistics of your data, but also the integration and management of that data into well-organized systems. Proper assessments of those fundamental systems and processes should be well understood before investing in an analytics offering. Timelines and implementation costs are most likely to go over job estimates in the data integration and management layers.
The unique technology, go-to-market strategies, and data readiness unknowns are potential complications, especially on large-scale projects. Vendors may say that in order to get the most value out of your analytics investment, you should implement the platform on all major production lines, sites, and levels of business. The risk inherent in large projects can cause companies to shy away from yet another big investment. Some might choose to perform just enough work in house that they miss out on potential returns from a comprehensive analytics platform.
Rather than not pulling the trigger on a large analytics investment and missing out on the benefits entirely, companies should consider focused, small-scale analytics pilots. There are many benefits to investing in a small pilot:
1. They validate success metrics. Small pilots allow you and your vendor to determine success criteria for the pilot. These metrics may be implementation date, number of models built, operators providing input, and other useful metrics.
2. You can fail small and fast. A small pilot reduces the risk of big, damaging failures as you try to integrate all systems. You will likely see failures in data standardization or connection issues across your network, but a small pilot makes these failures manageable.
3. You have time to assess vendor capabilities. A pilot allows the client to assess the vendor’s service ability and decide whether to continue the partnership on a larger scale.
Periodic wins do not build momentum; steady, measured progress in one direction builds momentum. Consider the same strategy when you are looking to invest in analytics.
Daniel Riley is an analytics service leader at Interstates, a certified member of the Control System Integrators Association (CSIA). For more information about Interstates, visit its profile on the CSIA Industrial Automation Exchange.
