Digital Transformation: People, Process, and Technology Mature with Better Technology, Methodology,

An issue in the manufacturing process needs a solution. It could be mechanical equipment reliability and integrity, or inadequate process equipment performance and resulting failures, or another complex issue. How do you select the solution? I remember years ago being told in no uncertain terms by a major oil company that the staff did not care how good the technology performs; if it cannot fit with the company skillsets and work processes, then it cannot be successful. Let’s look at what that means.

Technology, Methodology, Workflow

Any solution to manufacturing and production problems must involve an intimate interplay of the intrinsic technology, a methodology, and then the relevant workflow. Every solution needs all three. Like the instruments in the orchestra, led by the conductor, they must all play together in harmony to make music. But, the proficiency of each adds value to the benefits derived from a full solution.

I work in the Machine Learning space, a wide and massive field with, it seems like, dozens of venture capital-backed startups entering with all manner of proposed solutions. Here, in industry they all chase after reliability and maintenance which is a huge opportunity that has inspired so many to throw their hats into the ring.

Let’s go back to what’s needed to solve complex problems for companies in asset-intensive industries. The vendor’s application is never enough. An effective solution always requires an intimate interplay of the intrinsic technology, a dependable methodology, combined with a work process(es). A solution dominated by technology alone is too one-dimensional because it demands great skills, understanding and experience and likely manual, inconsistent methodology. Not fun, hard to do, takes a long time to do one project, and cannot adequately scale. Great technology isn’t enough. Has it ever been?

Hide the Complexity

An iPhone does all kinds of great technology “things” that you do not need to grapple with and it all happens automatically behind a simple-to-use screen where your finger is the pointer. Imagine a Machine Learning solution developed the same way – designed to stop machines breaking. The technology part might be insanely competent, but abstracted so that maintenance and operations people could use it to solve the problems that they understand. The methodology would be automated doing most of the work every few minutes, delivering accurate trustworthy predictions of the future at the speed of light. And the workflow process assures that you impart the right guidance simply and easily, by pick and choose, to focus the solution. But in return, it provides actionable information to resolve the impending breakdown with minimum disruption, and/or advises the immediate changes in operational procedures; what to do to correct the issue, to make the deterioration go away, and eliminate the breakdown and subsequent maintenance.

Bridge the Competency Gap

However, it is well known that in the oil and related sectors, close to 20% of the workforce must retire over the next few years. Chemical engineers take years of domain experience with them; expertise that quite literally ran the operation in the old days. Such knowledge is not easily, or cheaply replicated with new employees who lack the skills and experience. Compelling reasons demand solutions to help bridge the competency gap.

Layoffs in the energy industry after 2014 climbed to over 350,000 in two years, resulting in a shortage of new people to support the return to growth. Additionally, the World Economic Forum (WEF) estimates Industry 4.0 (IIoT) has potential to contribute $20 trillion plus to global GDP by 2020. But, WEF also advises in the WEF’s Accelerating Workforce Reskilling for the Fourth Industrial Revolution, June 2017, that “… 35% of the skills demanded for jobs across industries will change by 2020.” In the same vein in 2012, The Harvard Business Reviewperceptively warned in the Data Scientist, “The shortage of data scientists is becoming a serious constraint in some sectors.” Furthermore, McKinsey & Company weighed in on the talent crunch claiming that by 2018 (now!) that the United States will experience a shortage of 190,000 skilled data scientists, and 1.5 million managers and analysts capable of reaping actionable insights from the big data deluge,” and “… Industry is competing for big data experts/talent with other industries like medical, retail, finance, bio, and tech.” For manufacturing, we must also consider competition for data scientists from metals & mining, power & utilities, and transportation; all global industries with enormous economic interests in answers from analytics.

How do you select the solution?

The industry urgently needs a new path forward with solutions that overcome the prevailing realities. The sustainability of manufacturing business depends on it. The inventors of Mtell application predicted that a lack of data scientists would be a serious barrier to realizing the benefits of analytics – a reality reflected in a recent survey of AspenTech customers. Further, domain knowledge from process and mechanical subject matter experts must lead the data science, and both must fit seamlessly with a methodology that automates and simplifies the execution, binding into the right work process for the right circumstances. Only the correct interaction of the fundamentals will help the industry reach the highest levels of operational excellence and business sustainability now and in the future. The Aspen Mtell design foundation assures fast and easy deployment using the existing workforce skills, and a supportable and sustainable solution for the future. Such deep support for the intimate interplay of technology, methodology, and workflow creates a world that doesn’t break down.

Written by:  Michael Brooks, Senior Director, Business Consulting, Asset Performance Management (APM) Business Unit at AspenTech, for ARC Advisory Group blog.