Predictive Analytics Pilots That Work

Credit to Author: Guest Blogger| Date: Wed, 27 Feb 2019 16:36:43 +0000

According to Forbes, “Digital Transformation isn’t a buzzword anymore. It’s the way.” In today’s increasingly competitive marketplaces, every business is looking for an opportunity to exploit a new competitive advantage that helps combat their industry and market pressures. Numerous cutting-edge technologies promise to help overcome these pressures. With the adoption of digital transformation increasing across every industrial vertical, companies are looking to technologies like predictive analyticsmobility, and augmented and virtual reality to uncover those advantages. The question then becomes – how do companies implement these technologies with minimal risk? The answer – Pilot Programs.

A lot of the technology being introduced in industrial applications today may seem like magical black boxes that can only truly be understood and leveraged by data scientists. Take Predictive Analytics for example. While that term is thrown around quite a bit in industry publications, a lot of companies may be struggling with how this technology can be implemented in a way that proves its value with minimal upfront investment. This is the perfect case for a pilot project. The concern that companies and vendors need to watch out for however, is a term being referred to today as pilot purgatory.

Pilot Purgatory

Pilot purgatory may sound comical at first. But with all the hype and noise around digital transformation technologies like predictive analytics, machine learning and big data, pilots are often implemented at a snail’s pace as users get up to speed on how the technology works and explore how it’s applied to their specific use case. This can result in pilot programs that run for years instead of weeks or months. In fact a recent McKinsey report notes that for Industrie 4.0 pilot projects:

  • Only 30 percent of the pilots end up reaching scale across the entire organization with companies failing to capture value from 70 percent of their pilots
  • Some 85 percent of the companies surveyed spend more than one year in pilot mode, while 28 percent spend more than two years

To avoid pilot purgatory when it comes to predictive analytics software here are some things to consider.

Qualify Vendors Prior to Pilot Phase

The core competencies of predictive maintenance software vendors can be identified before the pilot process kicks off by evaluating the vendor’s:

  1. Existing customer base/references
  2. Software capabilities and features
  3. Proven software scalability
  4. Domain knowledge (for predictive analytics and asset performance management this goes beyond data scientists).
  5. Ability to act as a long term strategic partner with a roadmap for the future
  6. Support and services approach
  7. Relative cost of solution.
  8. Total cost of ownership (TCO).

A Foundation of Strategic Partnership

After numerous successful pilot programs involving AVEVA’s PRiSM Predictive Asset Analytics solution we’ve found that the most successful pilots are not the result of an attempt to uncover a business case but instead to validate the business case to invest in the software in the first place. The pilot project should focus on the vendor and customer working as closely as possible for the pilot to succeed.

Tips for a Successful Predictive Analytics Pilot Program

While there are a number factors that go into ensuring a successful pilot project when evaluating new technology, here a few tips to help guide your path.

  1. Recruit Executive Leadership Support. With any technology project it’s important to get buy-in from the executive leadership team. Some leading companies such as BASF have even developed specific teams focused on digital transformation.
  2. Define scope and success up front. Work with the vendor to identify specific and measurable outcomes that will result from implementing the technology. Again, the focus of the pilot should not be to define a business case but to instead prove a business case. For example, proving that the software can predict equipment failures that avoid $30+ million in costs.
  3. People are the key. At the end of the day people need to use the technology for it to be of any benefit. That means that putting a training program in place is mandatory. And if the pilot is successful, continuing to invest in training on a regular basis.
  4. Create a timeline. Pilot projects should not continue indefinitely. It’s critical that a timeline is attached to the pilot program that specifically states a go-no-go point in officially adopting the technology. If the software can’t prove it’s worth within that timeline then alternate solutions should be investigated.
  5. Formal Project Management. Pilot programs need the same level of attention as a production level implementation. Resources to conduct the pilot should be assigned accordingly with a documented project plan in place before the pilot kicks off.

future-oil-refinery-technologyFor additional tips on how to avoid potential failures when it comes to digital transformation and other technologies like predictive analytics, check out the Harvard Business Review article Why So Many High-Profile Digital Transformations Fail. And remember that when the pilot program is successful, a plan must be defined to scale and rollout the technology as part of an overall deployment strategy.

To learn more about how predictive analytics fits into an overall Asset Performance Management strategy view the webinar Implementing Predictive Analytics in an Asset Performance Management Strategy.

 

Matt Newton Senior Portfolio Marketing Manager

Written by:

Matt Netwon is a Senior Portfolio Marketing Manager at AVEVA. With over 15 years of experience in the technology sector as an applications and systems engineer, Matt has extensive experience in supporting embedded platforms, automation systems, wired and wireless networking, network security technologies, and the Industrial Internet of Things.

 

The post Predictive Analytics Pilots That Work appeared first on Schneider Electric Blog.

http://blog.schneider-electric.com/feed/