Predictive analysis is Key to Effective Decision Making, and Future of Industrial Operations

Last week I toured the east coast of Australia, engaging with Food and Beverage customers about opportunity. In Australia like US, the focus is on becoming competitive and effective to deliver high value products, in a more timely manner.

At the ARC conference in Orlando in Feb, we saw increased discussion around predictive analyst ices.

The whole objective is to move from the "as is" (which is traditional alarms) to the "to be" state, so the impact of time to detect, time to react to a conditions, has less impact on the situation, reducing cost, risk.

The diagram below I have shown before but is still one examples of why we must start a shift to predict through patterns, and relationship with between variables to see conditions developing before they develop into risk conditions.

As can be seen from the ARC diagram below most of the systems we build today are on now (dashboards) and past (Reports), but the edge is going come from the future, often learning from past to predict the future. The user must be looking to the near future to allow decisions this could minutes to 12 hours into the future with high predictability so he can ask “what ifs” and help make rapid decisions.

 All the areas in the orange provide new insights into the direction of the operations or their situation so they can predict, what you they do now. This requires lots of information, and analytics to predict models, and patterns.

Good example is the partner company Asset Insights from Pattern Discoveries Technologies.:

Taking known events around assets, combining these as context to determine patterns from within the historian data to bring value a repetitive pattern that can now be used to “shift to the To Be” state. So decisions are made faster and in more confidence, with insight into what to do next.

Software like this will grow fast around the core big data of such items as historians, alarms, and asset, energy systems.

Operations Innovation & Transformation – Training “Systems”

Blog post from Stan DeVries on his thoughts on the Training transformation, operational guidance.

Industrial operations are experiencing innovation and transformation in 2 aspects.  There is a pervasive move to unifying the industrial enterprise over multiple sites with a more holistic view, and there is a pervasive shift to operational teams that spread across the sites, centrally governed, where subject matter experts (SMEs) make use a timely and dynamic decision support system in a dynamic operational world.

Using the recent discussions in the MESA community about the recommendation to consider “behaviors” and “systems” as separate, where “systems” help to achieve, sustain and improve “behaviors”, there has been a revolution in manufacturing and operations Training Systems.  The major change is from a single user, focused on his/her span of control with an isolated context of process or machine control, to multiple users handling multiple processes or machines, and handling more than just speeds and pressures – they contribute to operations performance indicators such as quality and unit costs.  The “contract” between the enterprise and the operator has dramatically changed.  In addition, even the traditional narrow span of control isn’t sufficiently effective when operators have never experienced a full start-up or one of the more severe abnormal situations.

The transformation of the Training Systems and associated innovations is to a group experience training system.  For many operators, there will always be at least one training scenario with a role that is closer to the traditional “fixed span of control” that focuses on a smaller area of operations or set of equipment.  The 4 quadrants are summarized as follows:

An important aspect of training in these 4 quadrants as training for frequent change that is coordinated in teams.  The following diagram shows a “tram line” display.

It is important to note that in most industrial operations, this activity is governed by scheduling or “dispatching” as it is called in the power generation industry.  It is still important for operators to be trained to perform safely and effectively in this dynamic situation.  Using the process control metaphor, this is training for when the Setpoints change dynamically.  The transformation in training for the 4 quadrants is when Setpoints and Measurements change dynamically, in both scheduled and unscheduled situations.

The following are examples of training scenarios for each of the 4 quadrants:

•        “Fleet” management with a fixed span of control (lower left quadrant).  The trainee learns how to operate their normal area or group of equipment without exceeding quality, operations or safety limits, with dynamic changes in setpoints (setpoints are adjusted by the operator in accordance with instructions).  The flexibility in setpoint position is the key learning experience in this training.
•        “Fleet” management with adaptive span of control (combination of lower left and lower right quadrants).  The trainee learns how to switch between operating other areas or groups of equipment without exceeding quality, operations or safety limits, with dynamic changes in setpoints (setpoints are adjusted by the operator in accordance with instructions).  The flexibility in span of control is the key learning experience in this training.
•          “Fleet” management with adaptive span of control and governed changes in targets and limits (combination of lower left, upper right and lower right quadrants).  The trainee learns how to operate as part of a team, where flexibility and agility is increased as the team adapts to exploit opportunities or minimize undesired events.  Typical scenarios include rapid increases and decreases in throughput, but can also include rapid changes in quality (terminology varies in some of the customer industries).  The teamwork is the key learning experience in this training.
•           “Value chain” management with a fixed span of control (upper left quadrant).  The trainee learns how to adapt to a rapid change in equipment load (reduction indicates a slowdown downstream), where the setpoint instructions haven’t changed but the ability of automatic control to regulate the equipment has changed.  The teamwork is the key learning experience in this training.
•         “Value chain” management with adaptive span of control (combination of upper left and lower right quadrants).  The trainee learns how to switch between operating other areas or groups of equipment without exceeding quality, operations or safety limits, with dynamic changes in equipment loads (reduction indicates a slowdown downstream), where the setpoint instructions haven’t changed but the ability of automatic control to regulate the equipment has changed.  The teamwork is the key learning experience in this training.
•          “Value chain” management with adaptive span of control and governed changes in targets and limits (combination of upper left, upper right and lower right quadrants).  The trainee learns how to operate as part of a team, where flexibility and agility is increased as the team adapts to exploit opportunities or minimize undesired events.  Typical scenarios include rapid increases and decreases in throughput, but can also include rapid changes in quality (terminology varies in some of the customer industries).  The teamwork is the key learning experience in this training.

All of the above training scenarios maximize the dependence on sufficient training system fidelity and ideally the fidelity of the training experience.  The training experience includes collaboration and navigation of facilities and operator actions in the “field”.

How Far Away is That Operational Workspace Transformation Inflection Point?

Yes, we have known this for a while, yes we have heard of the mobile device outselling the PC desktop, but has it really hit home. Last week I was on tour on the eastern coast of Australia, and talking about the rapidly approach inflection point for the “Operational Work space Transformation”.

Source Mary Meeker from KPCD Presentation on Internet Trends   

The above diagram really cemented the dramatic change in the last 4 years in the human interface market from the point of operating systems. While this will not translate yet to the Industrial Workplace, already the trends of allowing My OWN Device onto the industrial work place is increasing or accepted practice. This has brought the iphone and Android devices into the industrial operating environment.

As was one comment last week many of the commercial stores do not assume actually expect that their orders are entered more on mobile devices than desktops. Already on some sites especially in the building/ facilities sector you can assume that a significant % of inputs and reading of information is from mobile devices, eg non Windows OS.

In some of the talks, last week people were surprised at the bluntness that a few of presenters, including myself were saying that the most significant Event since Y2K to effect the Operational Work space landscape and architectures is rapidly approaching. With the :

·         Changing profile of devices that industrial data / information has to be delivered to, from the traditional assumption of the 90s and 2000s of Windows as the default platform.

·         The changing demographic to Gen Y and the digital native, multi tasking engagement

·         The shift to teams vs individuals and collaboration and sharing to make rapid decisions

·         The fact that experts will be virtual and thus have increased responsibility and requiring increased access to trusted information in more real-time.

·         The need to agile so the operational work space must flexible to absorb change.

Is the inflection point next year, or the year after we do not know but it is coming, and fast, are we ready as both vendors and engineers and designers to deliver solutions to satisfy?

Certainly the investments are happening in Invensys/ Schneider software; the challenge will be the ability for the market to absorb this change, while maintaining operational continuity.

So why 3rd Generation MES , Model driven industrial Operations is the Only Path for Today's world

A lot of reactions to last 3 weeks of blogs, interest and discussion, but that discussion has been direct with me on email or face to face, pity as many you miss out.

The discussions reinforced that the time of model driven operational solutions is here and not for one reason but many, and that it really is the only way to develop an architecture that can grow in functionality and agility and size. As well as deal with the transformation of the operational workforce, and the new paradigm this will bring with the dynamic, short tenure ship.

Combine this with the agility to evolve, add differential value through the capture best operational practices, embedded them, and evolve them.

Time and time again CEOs and COOs say as they move to off the shelf products, their differentiation will be based on the operational practices that put into practice.

Last week I discussed the critical move away from customization in solutions to configuration, within the framework of the tools. A lot of the feedback, this week, was in agreement that their existing MES type solutions either custom( internal) or first or second generation MES with significant customization have worked well in a STABLE operational environment. Over and over again the comment came that it is time for change, to a system that can:

1/ Enable evolution of operational practices by capturing the best practices

2/Rapid product introduction and evolutions

3/ the ability to scale over multiple sites in a sustainable way with common product definitions across the sites.

4/ transparency across site and multi site eg the value chain

5/ empowerment of decisions in the NOW and consistency of actions across different roles, shifts

Two feedbacks this week relative to the modern operational system:

1/ MES functionality as defined by Mesa and ISA95 is a commodity

2/ assume operational/ production change 3/ assume operational workforce, people transition and evolution

These are key comments when we consider why the 3rd generation MES based upon core MES functionality in a scalable architecture naturally extended with model driven (workflow) operational practice capture.

So I get shocked when I hear that people are still saying develop a custom MES/ MOM system, why waste the energy on a mature technology, and focus your energy on differentiating through the capture and embedding of operational practices and actions, while empowering decisions in the NOW across the operational community in the organization.?

Third Generation Mom/MES with Model Driven, Provides the key to Avoiding Customization

In speaking with customers from multiple industries the desire for "operational management and manufacturing intelligence" grows, but so do the horror stories of the promise vs reality. I keep asking why the horror why the disappointment, and in a series of 36 interviews it became clear that it was a number factors:

1/ vendor solution miss fit

2/ too much customization came over and over again, as key pitfall 3/ customer expectation vs reality and perpetration

Too often people initially "point fingers" looking for the vendor or implementer, but the mature, realistic interviewees admitted it was a mixture of their own preparation, culture and expectation as well as the vendor and implementer partner.

Without about a common message came across avoid customization at all cost, two interviewees talked about how they were replacing a home grown system, and took a vendor product and customized it heavily for "product x" and found that they could not take it to the second and third plant and ended up with a system similar in functionality and cost to sustain as the original home grown system.

They both admitted they did not step back and look at the new operational processes they should apply for the next 10 years but instead based their design on what they had. Both are now reviewing a new approach for the other plants. This story in different levels of complexity was not uncommon, and many had first experienced this with the ERP system.

Another huge challenge is the different cultures and operational practices on the different sites, causing significant challenges in achieving consistency in product / service quality, and operational efficiency.

There were many people where confused and bitten from applying the wrong product to a particular industry. It is important to achieve minimized customization the modern MES/ MOM system must come from that industry segment e.g. you can look at:

1.       Discrete industries

2.       Hybrid industries

3.       Continuously industries

We in the industry cause much of the problem as MES is term applied across all three, but they very different, the key to a vendor/ packaged MES is to look at it’s origin, and that will indicate the model that it fits in the above 3 broad industry areas.

So when you look at 3^rd generation MES why it really has hope is that it is model driven, avoiding customization, through allowing managed configuration that will enable companies to evolve. With the changes in the operational processes this key, as this is where you gain the significant advantages. If you choose a correct MES from your industry segment, and then apply a model driven/ workflow approach to configuring your operational practices this provides the most advantage and flexibility. Combine this with object driven platform when connecting to an automated world for real-time events, the system can become proven quickly but most of maintainable and evolutionary.  

Key is uniqueness does not mean customized code, it should be sustainable configuration based on models that can used over sites, so operational practices (your advantage) are consistent but can evolve!

Reality in the Transformation in the Industrial Training and Knowledge Transfer, With a Power Plant, Why not You?

I have discussed at length the transition of our workforce and that it is not just going to a new generation ( 42% of the workforce by 2020) but it is the dynamic culture of this workforce ( tenure ship less that 2.4 years in the role by 2020) that poses one of the biggest challenges to the industrial operational/ automation workplace.

At a dinner in Florida I was sitting next to a gentlemen in charge of "change management" at a counties power facilities major power stations. A major part of the focus is on training, and knowledge transfer. The conversation initially focused on the challenge of "baby boomer" retiring over the next 5 to 8 years and the knowledge, experience "walking out" the door. It was interesting to understand how stable his  automation and operational work force has been for the last 20 to 30 years. They are a premier employer in the area, and most of the current work force has come from trainee and full careers there, many were there when the power plants were commissioned, growing up with the plants.

Reminding me of a saying by a colleague "that when he joined the workforce on plants, the average age was his age eg 20, now that he is in his late 50s the average age of the workforce on a plant he visits is still his age ( late 50s)." As we had dinner this, saying was ringing in my head as the stories from the power plants came out.

As the dinner progressed the other challengers came out that they were building new gas powered plants, shutting down some of the older power units, and the new workforce was aged less that 35 years old, (most younger) and little to no experience. The shorter tenure ship in a role he had not seen yet, but when discussing he had noticed a trend to rotating roles, and want to progress, new challenges, and with the new gas plants with new technology and learning he saw the challenge not far off.

So he has embarked on a significant Operator Training Simulation Program. Work on three fronts:

1/ power unit operations

2/ power unit maintenance and start up, shut down

3/ environmental / energy management

These are 3 different communities of the workforce, the third being a specialized one which is becoming critical due to government regulations and the heavy fines, so understanding how to seal with different situations including weather and load is key.

The first step is to really structure the training knowledge capture from the “baby boomers" capturing their experience in operations reacting to situations, and performing maintenance. Action and knowledge they take for granted now caught in the simulation models and built in as training scenarios that can play over and over again.

His learnings are also supported in the diagram below:

He also talked about not just simulations, but his learning has been the other training tools, measurements within the OTS tools that he did not initially appreciate (or use until recently) that allow the training to evolve, and improve as well develop the team.

The big walk away I had from this dinner discussion was that he could not see how they would go through this workforce transition, and plant transition, and maintain the current operational levels in efficiency and safety without automated, simulation tools.

His whole story was micro example of what I am seeing across the world, but so many are not acting, he had been implementing a "change management" plan based around simulation learning for the last 4 years, and he was still evolving, but seeing significant returns.

On the new plants, they will have simulation as part of the plant design and operator training simulation from the outset.

Yet a week later across the other side of the world in mining, (where an even more dynamic workforce transition is happening,  with combination significant workforce rotation and retirement) the shift to simulated online training is seen as a "nice to have".

Key as we look forward there are 3 big areas that need to align, Operational performance Asset utilization / efficiency Operational workforce effectiveness. They are all equal and dependent on  each other, and the leading companies will be the ones that master the balance across the three as the dynamics in all three increases.

Next week I will expand on knowledge dissemination, learning transformation that is and will have to happen in the industrial space.