Forecasting and Predicting, Must be a Cornerstone of the Modern Operational System

For the last couple of weeks Stan and I have been working with a number of leading companies in Oil and Gas, Mining, and F & B around their Operational Landscape or experience of the future.

Too often the conversations start off from a technology point, and we spend the initial couple of days trying to swing the conversation to the way in which they need to operate in the future and what their plans are around operations.

It becomes clear very quickly that there is allot of good intent, but real thought into how they need to operate in order to meet production expectations both in products and margin has not been worked through.

Over and over again we see the need for faster decisions, in a changing agile world, and this requires an "understanding of the future" this maybe only 1/2 hour. The time span of future required for decisions depends on role, (same as history) but it is clear that modeling of future is not just something for the planner, it is will become a native part of all operational systems.

This blog from Stan captures some of the necessary concepts.

  

Operations management systems must deliver better orientation than traditional reporting or decision support systems.  One important aspect of operations is the dynamic nature – there will be a journey of changing schedules, changing raw material capabilities, changing product requirements and changing equipment or process capabilities.

It might be helpful to consider desired and undesired conditions, using the analogy of driving a car on a long trip.  The planned route has turns, and it may involve traffic jams, detours, poor visibility due to heavy rain or fog; the driver and the car must stop periodically; and the driver may receive a telephone call to modify the route.  The following diagram is a sketch which displays how an integrated view might appear:

In the above example, the actual performance is at the upper limit for the target, and the scheduled target and constraints will shift upward in the near future.  The constraint is currently much higher than the scheduled target limits, but it is forecast-ed to change so that in some conditions in the future, the constraint will not allow some ranges of targets and limits.  This simple view shows a single operations measure with its associated constraints and target.

• At this stage, we propose a definition of “forecasting”: a future trend which is a series of pairs of information, where the pairs include a value and a time.  The accuracy of the values will be poorer as the time increases, but the direction of the trend (trending down or up, or cycling) and the values in the near future are sufficiently useful.
• In contrast, “predicting” is an estimate that a recognized event will likely happen in the future, but the timing is uncertain.  This is useful for understanding “imminent” failures.

The following diagram shows an example of estimating the probabilities of 5 failure categories, where the first (rotor thermal expansion) is the most likely.

Given these two definitions, it is helpful to consider industrial equipment behaviors.

• Several types of equipment, especially fixed equipment such as heat exchangers, chillers, fired heaters etc. exhibit a gradual reduction in efficiency or capacity, or exhibit varying capability depending upon ambient temperature and the temperature of the heat transfer fluid (e.g. steam, hot oil, chilled water).  While the performance is changing, the equipment hasn’t failed, although its performance might reach a level which justifies an overhaul.  In extreme cases, sudden failures can occur, such as tube rupture or complete blockage.  These benefit from “forecasting”.
• Other types of equipment, such as agitators, pumps, turbines, compressors etc. exhibit sudden failures.  These benefit from “predicting”.

One analogy of incorporating both “forecasting” and “predicting” is that it is like driving a car without looking forward through the windshield/windscreen, such as shown in the following sketch:

In the above sketch, the road behind the car is clear, but ahead, a potential collision will occur.  High-performance operations requires that teams prevent unplanned shutdowns or other events.

How will we work in 2025?

During the holiday break I was catching up on reading, validating ideas, and directions, and I found this article on "Why we would work in 2020? from NASA IT Talk.
http://www.nasa.gov/sites/default/files/files/IT-Talk_July2014.pdf

What interested me was you had a big semi government organization often not know for agility, talking agility of missions, of different sizes, and also the workspace transformation technologies and experiences were the same as we are predicting in Industrial/ manufacturing operational space.
The link above takes you to the article and here is an extraction: The targeted outcomes are aligned as well, with agility, collaboration, understanding the future,

"As IT professionals, we are used to rapid changes. But compared to what’s coming, we ain’t seen nothin’yet. Of course, no one actually knows the future, but by predicting it, we can make better decisions today that will help us become more effective tomorrow. The purpose of this article is to start a discussion so we can innovate together to help NASA IT lead the way and prepare for how NASA employees will work effectively in 2025. It has been said that the best way to predict the future is to create it. While we may not be able to create the IT future by ourselves, we can certainly in‑fluence it. A good way to accomplish this is to:

(1) collaboratively predict the future;

(2) test it together now with leading industry innovators by creating meaningful and evocative prototypes that provide high value for our constituents in the NASA environment;

(3) measure the results

(4) communicate the results as visibly and loudly as we can.

So, what will the technology environment look like in 2025, you ask? OK, here’s a prediction at a subset of the new normal in 2025:

• ·         3D printing / scanning / faxing is mainstream.
• ·         Consumer robotics is everywhere and really cheap.
• ·         All data is accessible, searchable and usable from any device.
• ·         We can use unlimited computing and storage through cloud computing.
• ·         Computing is wearable with any data accessible at any time.
• ·         Reality is augmented via modeling by default through our mobile apps and wearable computing.
• ·         Space is partly commercialized and NASA routinely partners with commercial and nontraditional
• ·         entities.
• ·         Over 10% of cars are self-driving.
• ·         More than 50% of  employees are Millennials.
• ·         NASA looks and feels much more like a startup than we did in 2014 and we use
• ·         crowd sourcing routinely.
• ·         Projects are accomplished in months, not years.

How will we work?

·         We will routinely use effective, rapid prototyping with faster, lighter, cheaper, and more effective infusion of the latest technologies into the NASA missions. Agile development will seem cumbersome in comparison.

·         We will evaluate and use the most effective emerging tools as part of our normal work. Visual programming and modeling will be expected and NASA will show visible leadership to industry.

Where will we work?

Simply put, NASA will be the workplace of choice. We will have a balanced, “startup-like” environment with mobile, reconfigurable, ­t-to-purpose workspace that enhances personal productivity and job satisfaction.

Working from anywhere with any data and any device will be the new normal.

Who will perform the work? NASA will be the employer of choice and the

partner of choice for the next generation of startups, industry, partners, and competitors. What about “the crowd” you say? Bring it on! Crowd ideation / development / funding will be commonplace and highly effective.

What will we work on? We will be equally adept at small and large missions, for both wild and feasible ideas. We will use industry for transportation. We will be on our way to 3D printing on Mars in preparation for sending humans to Mars. Asteroids will be within our grasp (literally). Submarining under the ice of Europa will be imminent.

We will monitor and protect our planet with millions of sensors composed of official NASA instruments and crowd-sourced wearable computing and nanosats. And that’s just a start.

Here is a sampling of predicted changes and prime candidates for prototyping that will show us the way to taste test this future now across NASA Centers and with leading industry innovators:

• By taking advantage of Big Data and Analytics, we can easily ­nd, store, share, and update all relevant information when we need it. We will provide self-service analytics to all who need it, so our decisions are based on data, not anecdotes. Robotic devices and scripts will collect valuable data for us 24 hours per day, every day.

• The Internet of Things and Wearable Computing will help us to have instant access to all this information at our fi­ngertips, on our wrists, in our glasses, via hand gestures, and by simply speaking the questions.

• We will use just-in-time training through videos created by current NASA specialists, and through specialized Massive Open Online Courses, all available from anywhere and any device via on-demand video snippets delivered directly to our favorite devices, such as smart glasses.

• 3D Printing/Scanning/Copying/Faxing will be mainstream and will allow us to hold effective

brainstorming sessions where we mix virtual and physical models regardless of where we are located.

Is this too Pollyanna’ish for you? Too conservative? Either way, please participate in the conversation and help us steer this train in the right direction, because it is already moving and speeding up, with or without us. Our destination is exciting indeed. And it’s all enabled by IT. "

We should not be surprised, but it is good to see validation of our thoughts.

Real time information Platform vs. traditional historian, Why it is Key to Pushing “Actionable Decisions”, foundational to the “Industrial Internet of Things” and Empowering the Teams.

Again last week I was presenting to a set industrial companies in water, food, and mining, and the topic of a "real time information platform" many questions.

My immediate answer is “what are you trying to do? " " who are the users targeted to interact with the system, and what decisions and actions are they expected to take?". These last two questions usually leave a complex blank expression on people's faces.

Many are engineers who have been asked to investigate, and they centered on the traditional approach of a "data centric" historian centered  world, leading with a technology strategy. The question of what people will use the data for, what roles and actions to be taken are secondary in their minds! 

Why is this when if someone had wanted a "historian" they would have asked for it. So why a platform, what does real time mean, and key is information.

It all comes back to one of the quadrants we talk about in the "operational transformation" around networking a series of assets, plants into a a "trusted" information system. That "actionable decisions" can be taken by a ever increasing community of operational people across the operational landscape.

To me it is understanding this community of consumers and what their requirements, uses are is key:

1. What activities, decisions, and actions they are expected to take?
2.  Their roles, skills, and approach is their time frame, location relative to the data
3. Their context and understanding of the plant, asset or process in question, as their is a growing trend of highly educated skilled people on assets, process. With little or no practical experience on the asset, and more than likely will not have visited site.

On investigations you find you have the traditional process engineers, who need the trend analysis and discovery of potential improvements. 

However, there is a growing tribe of people who need to make actionable operational decisions. They will not monitor the system must best "self-aware, and living" (exception based) capture the data, transformation  it into information.  Apply experience and knowledge, clear understanding of the situation, and what are typical actions with "best operational process" provided to take action.

This is very different to everything getting data stored and then extracted, yes in this new world there is history as it provides the history for reliable knowledge and basis for wisdom or " application knowledge".

The real key is the change in approach from “predictive to prescriptive” which embeds the “actionable decisions into the model. Empowering the operational team, no matter the location or experience with decisions and associated actions.

Understanding this maturity curve and evolution is what we see as foundational to the success of “industrial Internet of Things”. Through the embedded practices provides a basis for the changing workforce to act and make decisions in a timely manner.

However, these two communities in the industrial landscape are interlocked for success. The two communities are:

• Community 1: Process, performance, optimization team that accesses the data with trending, analysis, and predictive tools. Identifying the trends, conditions by applying their experience combined with “big data” techniques allows these conditions, to be seen in the “to be state”. If captured in a managed configuration framework, that will allow roll-out over sites and sustainable evolution. These become embedded into the system, for adoption by the operational team.

• Community 2: Operational Team: This is the dynamic team, from roaming people on the plant to central operational teams, to virtual expert teams, collaborating together in real time to enable “actionable decisions” no matter role, location, and experience.

The diagram below shows the this maturity of capturing this “applied knowledge” as Managed “Actionable Discussions” that interact with people, assets and process as key, very different a traditional historian approach.

The “Real Time Information Platform” provides a real-time "living" model that is self-aware that captures validates the data with rules aware of it is current state. Storing this data in context and rules and calculations in that provide motivation, embedded operational process, and awareness to correct people. Fundamental is the "trust" worthiness of the information, without impacting current automation systems. The ability to have sustainable evolution and scalability, through managed components that represent the assets and processes (actionable decisions) to the model is available on storage side in history and real-time.

You cannot do this with Historian (data centric) architecture and solution. Make sure you looked at who the communities of users you are satisfying now and in the immediate future?

The Change Landscape of business Intelligence, EMI and Analytics driven by Information Driven Companies

All through this week in Europe I have spoken with customers looking for decisions in the NOW, and more people empowered to make these decisions. This does mean the traditional worker is evolving to knowledge worker, and this is across the different roles in the operational plant. But following on from last week’s blog the drive for realtime decisions is driving up the requirement for more advanced analytics to enable that decision, especially as the experience and time in the role of the decision maker reducers.

This drive for information and decisions, is causing significant other transformations in intelligence segment, and these are captured in a recent ARC table:

Some of the fundamental concepts it points out:

• The shift in time focus from Past to Future: in the industrial world I would put that as truly an expansion from the past and current to now past, current and future.
• From a performance to a predictive view that enables that decision
• Move from Batch data to realtime, this is key as we move from reports to dashboards that are dynamic with small trend tails showing now and immediate past easy to understand from a glance.
• Move from IT intensive creation to Self service: this is even more apparent in the industrial space as operations want to be self empowered without the complexity and delay in engaging either engineering or IT.
• Users move from a few Gurus to a collective team of people what I referred to as the flexible operational team (many blogs on this) where experience is shared to achieve realtime decisions. Virtual experts in an active community either from within a company across sites, and subcontractors/ suppliers can now be in the realtime decision with the on plant person.
• Deployment will shift from “on Premise” to a “Managed set of on demand services” this while only just starting in the industrial space makes logical and even more sense in the industrial space. Because the information and these virtual communities will live outside of plants and across the world. Companies are talking of Asset Facebook concepts to where a community of peers and experts across plants working on similar equipment and processes can interact share and make informed decisions together. The concept of a hosted set of services for an Information environment will be foundational for this to work.
• Another comment is the automated actions, I see this more in the industrial space as the shift from just supplying information to having embedded operational procedures to guide users through consistent actions.

Yes,  the traditional EMI (Enterprise Manufacturing Intelligence) system will transform dramatically, and it will need to work well with the I tools which are also transforming, as well as provide “preconfigured experiences” to enable users to answer known decisions and enable “self Service” for rapid adoption.   

 

Decisions in the NOW, increases the desire for analytics as a key component of an Information Driven Enterprise

Again on a flight to Europe this week, I struck up a discussion with a fellow traveler who is out the oil and Gas industry while sitting in Abu Dhabi on a layover. It was around the transformation of decisions support from reports, to dashboards, and now for a need for more real time decision support.  This discussion aligns with the Information Driven concepts and the transformation from reports to predictive and decision dashboards.

Information Driven Companies are looking for more than what has happened they driving to understand what will happen?
It is crucial to note Excel, BI and EMI tools provide extremely solid basis for analysis in the past and now, and in a focused area, but as decisions become more predictive the way to sure up the prediction is to start looking across significantly bigger data to see common patterns. EMI provides dashboards and alerts based upon basic rules and KPIs this is still required.

Source ARC

The above diagram shows the evolution of Enterprise Intelligence and Business Intelligence from the understanding of today to a more predictive requirement, this is from ARC. When I talk with customers, I use the Operational Excellence Journey diagram below to describe the evolution to real time, agile decisions. It is vital that companies accept that achieving operational excellence is a journey not a one off project as it evolves as the company learns and tunes.

As you move to right the analysis and analytics start looking for patterns, and relationships across data sources, and linking causes to a set of conditions. BI and EMI remain indispensable tools of information driven companies, but they do have limitations. For example, BI involves the IT and runs in batches of set time breaks while information driven companies are requiring broad access to analytical information and they need it continuously in real time. Finally to be able to adapt quickly to market place changes, information driven companies need to look forward, predicting what will happen next. Traditionally BI/ EMI systems have not incorporated predictive analytics tools to apply pattern matching rule and algorithms to historical data.

These requirements combine with the new technologies that are now coming common place and transforming the capabilities of large data analysis. Four overarching trends are transforming the industry: e.g.| Data, predictive analytics, self-service/ embedded analytics and cloud based analytics. Advance techniques such as data mining, predictive analytics, statistical analysis, data visualization, text analytics/ natural language processing can all be applied with e.g.| Data to discover new patterns and relationships opening new understanding and potentially operational advantage.. This significant trend, reinforced by the fact that modern predictive analytics tools do not necessarily require advanced skills, and thus overcome many of restrictions of traditional predictive tools. Many of us are evolving technologies and tools, that will analytics and simulation module as part of a supervisory/ operational experience (similar to alarming). Enabling small forward-looking models to run off existing  systems and history to allow a forward look based on the situation today. Combine this with users now getting information for decisions via advanced analytics tools on top of traditional data sources that they can use themselves(self-service) and immediate value. Next week I will expand on some of key transformations in the Intelligence BI worlds that apply in operations.