Big Data requires Pattern Awareness to Provide Situational Awareness

2012 saw the rise in what I call “Industrial Information Systems, Projects”. You may say “rubbish”, the whole historian, and information business has been around for years, and the answer is true. Today these projects are different dealing with a “lake of data”, delivering to more people, of different roles typically not even aware of what an historian is drawing data from many different sources including historians, xml files, transactional data sources such as MES and Batch systems, alarm, event systems, MS Excel and customer odd databases, as well real-time data. There is no one supplier, one source, or structure to this data. The challenge is when the context and knowledge of the data is retiring from the companies, but the size of internal community of roles and workers requiring access is increasing. How often I have been asked and discussed the issue of data validation and data awareness, vs architecture, and technologies at tossed into conversation hoping for a “silver bullet”, but I believe the solution comes with new capabilities like “Big Data” but also evolutions in existing industrial implementations, with a more holistic design!

I believe the growth will accelerate in “Industrial Information Systems, Projects” during 2013, and beyond, but this is not about delivering reports and information, it is about “empowering” the increased community in  business in making real-time decisions, based on real-time trustworthy, effective industrial information, no matter their location. I continue to get surprised by the notion that the solution is an Enterprise Historian on top of the existing system, acting as a data warehouse. I ‘Scratch my head” and usually “ask how to you know the data is valid, in context, and comparable. Too often it is a blank look they were lead to believe the data in or supplied from their SCADA, lower level historians, etc. is the only thing they need to access. Key to understanding is the ability to detect patterns, across data, but there needs to be enough context to allow the evolving big data tools to enable detection of patterns.

 

Last week this blog discussed exception based “self aware” models required in today’s proactive operational/ supervisory systems, especially as devices grow in intelligence capability. This model is also key to putting things in context enough to enable this analysis and patterns to be seen way further than process analysis. Big data concepts of pattern analysis, save that pattern, and now have it as auto detect on a similar pattern happening again, triggering an operational process that will continue a proven procedure to resolution, guiding the workers involved interacting in a consistent and pro active manner with the objective for early detection and fast resolution. This automatic pattern recognition, detection, and embedded procedure are one key aspect of the modern situational awareness concept. Building on last week’s blog concepts of the “self aware” model, if these smart devices and processes include a pattern recognition capability as part of the “self” intelligence, the shift is a response from the “as is” status to the ‘to be”. The diagram below shows a significant opportunity for improvement with the two “value of early corrective action” lines  effectively illustrating the value gain in early detection and pro active correct action.

The diagram illustrates how a condition over time “x axis” changes in cost/ value through time, and how the traditional alarm systems are in the “as is” state, and the whole objective to “SITUATIONAL AWARENESS” is to shift to the “to be “ state.

Why monitoring is, giving away to exception based ”Self Aware” philosophies?

Over a year, ago I jointly authored a whitepaper with leading thought leaders at Shell on the future of oil and gas field systems

“Establishing a Digital Oil Field data architecture suitable for current and foreseeable business requirements”.

The paper raised a number of concepts one of them, the need to move away from monitoring to exception based philosophies in the operational system. Over the last 2 weeks discussing with industry thought leaders in oil and gas, power, infrastructure and mining, the growth in data from the field through smart devices is accelerating the shift to a “self aware” / exception based systems. It has been fascinating to see how in a year the increased realization of a different approach to operational experiences to be able to enable effective decisions and actions. As the shift to exception base systems brings in concepts such as:

• “Self Aware” entities that will either be in the smart device, or higher level if it a smart process that can detect conditions and trigger notifications and operational procedures and awareness
• Advanced Process Graphics: the shift to uncomplicated view and easily identification.
• Situational Awareness: the ability to focus understand, associate related knowledge to rapid decisions and effective actions.

The paper centered on Digital Fields, but the principles apply across industries:
“Another key instrumentation requirement is to report by exception, i.e. Sensors to have a remotely configurable ability to detect and report changes. This approach will minimize source data flows and as much as possible distribute intelligence to the lowest possible level and thus minimize data volumes/complexity in PAS and higher level systems.

For example, consider a well head pressure transmitter. With this approach,  the transmitter will be smart enough to recognize and report only changes greater than say one percent, however, an authorized user should be able to remotely change the reporting threshold, to say .5% if so required. Compare this to the current approach where all data flows through PAS systems to historians which, ironically, store large volumes by filtering data in a similar manner. Hence the virtue of filtering at source, minimizing data transfer volumes and minimizing data storage in higher level systems. Note, exception reporting at source applies to analog and vector/matrix parameters - digital parameters naturally report by exception.”

The above extract from the paper outlines the concept, and why, the essential item is that monitoring is not practical as the data levels exponentially increase from the field. We need to reduce data on networks especially over distributed networks, and by going to local detection and “self analysis” will do this. The “self aware” approach will enable local detection of a condition that is escalated to operational people who can drill down, and draw on local data as needed. This approach also means intelligence is local or close to the source. How real is this? Last week an opportunity came to me where a gas well head will have all it is instrumentation and control on the well, and there will be a web server/ service on this well, and there will be a wireless 3G connection. A follow on discussion was at what level in the architecture would this detection be, it should be as a low as possible, down in instrument, controllers, but also discussion about “smart/ intelligent” processes, and process units deployed at the Operational/ supervisory platform layer of the architecture bringing together information, data from multiple sources and taking exception.

The above diagram shows the concept of smart process well, more than a data structure, and why a data centric architecture on historians will not be satisfactory. There is a need for an “Operational Application Platform” that enables smart “living” entities that can be distributed across an industrial landscape, and managed as standards. Feeding into local data storage systems, but doing more, by trigger exceptions, events, and executing escalation and awareness across the operational team, so enable alignment and responsiveness in this every growing operational environment.

I will expand on the "Operational Application Platform" vs the "Enterprise/ Data centric Historian " strategies both are valid, but in different uses, and there seems to be some confusion.

Achieving Operational Effectivness through Planned Operational Work vs Adhoc Work

As one customer said “We have teams and programs on captial asset performance for years, this will continue, but the real opportunity for gains is with increased efficiency across the operational team!”

 

The emergence of “Operational Performance Teams” to work along side the well established “Capital Asset Performance Teams”, with the focus on improving the operational effectiveness of the plants. In today's challenge is how do evolve the workers operational practises into a sustainable model. Moving to planned work vs ad-hoc work in a day and week takes a change in culture but delivers significant results in reliability, improved safety. The key to increased agility is through effective, real Operational Teams, having to collaborate and execute work in a timely manner, with the freedom to pass work items between team members for effective execution.

 

To achieve operational effectiveness the discussion between companies is not just information awareness, it is also looking at implementing operational work item systems that are not just logging and generating end of shift log reports. The discussion has shifted to a “journey of operational alignment and culture” as seen the below. Today many companies find they do not actually know what work items workers are doing in a day, many state that they ad-hoc and fire fighting, which is decidedly in efficient in time and cost, plus directly increases the safety risk. Companies have quoted figures of 35% of a typical plant workers day is planned rest is “unplanned”. The diagram above is intriguing depiction of a possible journey in a workers operational work management, key will be changing the culture of work management on the plant and dynamic workers.

The ISA 95 working group has also released a work management structure for operational systems as seen below:

The key are seven items that are required for Operational Effectiveness:

 

This is decidedly different to traditional electronic log systems, and the key is the fact that this is not a logging and task management, it is also capturing, and planning and assigned embedded operational procedures. The ability to pass work items between different operational team workers. On the plant we do often have a good work management in the Asset Management system, this is extremely effective for asset management, but the Operational Work management is for all items in the industrial landscape, and day. It is also targets improvement by performance monitoring of operational procedure execution and seeing how to improve these procedures. This can only happen if the actions taken are consistent, and this depends on “embedded operational procedures/ workflows”. Combining this with planning and realtime shift dispatching of work items, and constant realtime execution, transferring of work items made up tasks between team members with clear ownership so “accountability” which is key.

Invensys is continuing to evolve it is offering in the space to satisfy the every growing requirement of managing operation work items and therefore the efficiency of the human assets. Working with leading companies as they evolve on this journey of operational effectiveness, expect to see a lot more discussion on this area. My question is how can companies run a team without Operational work items management, that is why electronic log books ave grown in the last few years, but this is only a step in the correct direction.

 

 

“The Future is Here, It is just Unevenly Distributed!”

This statement caught my eye when reading and listening to feedback from this year’s ARC Conference 2 weeks ago. It came from a presentation by John Carey, Vice President of the Aviation, Industrial, Marine, and Energy Business at BP Castrol. Who followed up with another statement that got my interest, and is so true in what I am seeing:

"What keeps me awake at night is that everything that has made my organization successful today will block our success tomorrow."

What a true statement! I just wonder how many people understand it. At a dinner conversation this week I was in a debate on the changing world, and it was clear to all of us that this is not just a technology changing time, but there are three factors all playing at one:

·    Technology is evolving in leaps and bounds, but the role of the internet has had some impact on the industrial world, but nothing compared to what it will have in the next 10 years, where it will change the whole nature of industrial architectures. This is not due to the internet, it is actually due to increased bandwidth of infrastructure and high availability that enables the internet to now leveraged as natural member of the industrial architecture, instead of traditionally for an offline access for information, and basic email, notification capability. The whole ability to share, reliably put systems in the “cloud” and depend on them will enable the designs to accommodate the changing behaviors in the modern business. Mobility plays into this, with the explosion in devices and access, we can now naturally work and act while in a roaming world.

·    Globalization of Business ( “Flat World”): You may say that is old, yes it has been a concept around for a few years, but it is still taking hold but is accelerating at a rapid rate. Chaning the behavior in searching for products, how we buy, how we live our lives, we no longer restrained buy county of regional boundaries, we travel at whim, we communicate, and buy across the world. We all behave in business in a global virtual world. There is not a day goes by when I would not be on at least one meeting where we minimum of 3 continents in the meeting virtually, this is why we don’t work a 9 to 5 day anymore. If businesses are to stay competitive the “holistic” global approach to value supply chain and making it flexible and agile is key, and this drives core behavioral changes.

·    Cultural Shift to Digital Native: enables culture and thinking, driving shorter times in roles, average of 6 careers in a working life, (not counting jobs). The ability multi task, search and filter collaborate with people in more active community vs the traditional day in the life of workers 10 to 15 years ago. This is not a technology change it is a fundamental behavioral/ approach and cultural change form the “baby boomer and 1^st half Gen x” to those who were born past 1970 who execute a day totally differently.

Not often in history do you get 3 significant currents of change happen at once with each of the three effecting and enabling the successful passage of change in the other, which will complete the significant transformation in the way operate in the business and industrial worlds by 2020 and 2025.

Mr. Carey discussed both current megatrends and specific technology trends that are already shaping the future of manufacturing. Megatrends include:

• Changing demographics and lifestyles

• Emerging networks of trust

• Hyper-personalization

• Digital everywhere

• Sustainability

According to Mr. Carey, today's technology trends in communications, transport, processes, energy, and materials are increasing both in terms of their globalization and complexity. “To succeed in this type of environment, manufacturers must innovate, but rather than actually being innovative in everything we do, we often talk about innovation as a separate department. Manufacturers also tend to manage their businesses from the inside out. The real issues are the discontinuities. How can you get your hands around them if you focus internally? Suppliers and customers must work together in strategic relationships. This requires deep trust."  "Today, the customer is king,” said Carey. "This disrupts everything: manufacturing, distribution, and supply chain." In the 20th Century, manufacturers simply designed, manufactured, and marketed their products using the “fire hose” approach. In the 21st Century, however, product design, manufacture, and distribution is increasingly being driven by the customer and his/her individual preferences and requirements.”

 

Stepping back this is not a shock we have seen this in the car manufacturing, but of interest is the rapid transition from managing the process to manufacture goods to “managing the product and that intern manages the processes”. By focusing on the product and value to the customer including quality, satisfaction and timely deliver of value to customer, the competitive position grows. So is born the requirement of “Flexible Operations”.  This is not a “nice to have” it will be a requirement to be competitive in this global market. This can be achieved through an aligned business from business strategy, through operations to process control. Aligned operational teams that span the value chain that enable realtime decisions.

A comment in the ARC conference in Orlando in Feb 2013 was:

"To meet 21st Century demands for mass customization, factories will have to become more intelligent and flexible. "In China, they are not just building more efficient factories, they're building flexible factories," Carey warned." For many leading executives, they feel they have reasonable control over their fixed capital assets not surprising as this mature, but what keeps them awake is the “operational team” the human asset alignment, combined with the agility required surviving in a global “flat “world.

In the mega trends, Carey talked about “sustainability” this is about sustaining the planet, with discussion at ARC conference supported the impact and importance of this moral trend. Yes, a moral trend that driven by customers and market beyond government regulation that consumer (which is the primary focus) has a growing desire to choose a service or product based upon capability and the one that is doing the most on the moral issue of sustainability. This means zero waste and zero emissions both in the manufacturing process and across the total lifecycle of finished products.

As we have discussed a lot in the last year and will no doubt continue, this is not just a technology evolution, it more of a behavioral, cultural evolution that will force a significant change in the way design operational and industrial systems.

Looking out at the industrial landscape, the future has arrived in some parts of the world relative to generation Y and rotating worker (e.g.| China, South Africa, India etc), for some industries many technologies are already here in techniques and materials etc, but not in their industry, yes people have a look outside their industry to see how to achieve the future. Example is the transport industry which leads the dynamic operational centers vs the mining industry that is just starting on that journey. Also in many cases the global business landscape of the future is here and is accelerating in adoption.  

Many companies are well down this evolutionary path changing the operational systems to align with the new business and cultural behavior, but others are approaching the challengers in the traditional automation thinking. It was encouraging to see these topics come to the surface at the ARC Conference in Orlando and be debated, it shows what I am seeing the growing recognition by executives of the extend of the wave change we going to have ride, over the next 10 years.

Expanding from a Process/ Plant to Holistic Landscape will fuel the debate of “on premise vs off premise”!

I had an an editor speak to me this week about “the shift to Multi Plant view across the industrial landscape, then he mentioned cloud is a dream in Industry”. Quickly correcting him that it should not be a shift to multi plant view versus an expansion on top of the existing local/ remote operational view, and “cloud” is truly real especially with virtual teams, and multi site. This confusion is common at the moment with some vendors and consultants pushing the Enterprise Historian view that really focuses on only one perpective in operations, that of the analyst performance team. That does not mean there is a swing to a new architecture, the architecture must support at least the two communities of interest with different views:

• Operational day to day teams: requiring a view focused on limited time span e.g.| 7 days and view one process or plant
• Planning, Management, Performance / optimization teams and expert support teams : wants to look at broader view across sites, and broader time span of months and years usually offline for analysis, and strategic thinking.

The expansion to holistic view will also push the debate of the industrial architecture, to a combination of:

• On Premise computing: The local plants with real time needs and short time spans will often have the required information and capability local on premise. This can apply also apply in a remote operational center infrastructure which is still local to the remote site, or on the site and just virtualized to the center.
• Off Premise computing: The shifting of data, analysis, and actions to an off premise computing, either specialised data center, or “cloud” etc, private or public, freeing up the local operational day to day infrastructure from this heavily lifting. Providing also the ideal opportunity merge and consolidate a view and capability across the industrial landscape of a company, this will involve storage, but also realtime model of the production, and operational awareness activities across the different teams.

In the blog on 2020 the following characteristics were expected:

· One in three employees will be working casual, part-time or project base vs career in a company.

· The average tenure in a job / role will be 2.4 years

· One in four workers will be working remotely and virtually.

The ability to see information and data in a near realtime, view according to my focus is key no matter what current device or location being used. The ability to connect, collaborate and share advice is key to decisions in the now and the ability share knowledge when the average experience in a role will be less than 2 years The holistic view will drive many parts of the architecture “off premise” into a managed service.

Three times in a week one in mining two in food, from companies across the world, I was asked to comment on an “End to End” view of information, and especially state. One for operational effectiveness and agility, the other for reduction in delayed time in achieving a “positive release” of products to customers. In both cases the company IT and thought leaders raised both “cloud” and “big data “ techniques. As I mentioned last week, the acceleration in acceptance of these technologies is astounding in the last 6 months.

The mixed reaction to last week blog many agreeing, many debating the reality, but for the second camp, I think they unaware of cost, technology updates, and the expectation of the new generation for doing things fast. Also, there is a shift to be able to buy this holistic capability as  services  vs total up front.

I fundamentally believe that, within 2,  years,  we will laugh at this debate of “on premise” vs “off premise” in the industrial landscape, the modern aligned and agile business will have  an industrial/ operational architecture that naturally applies both On and Off premise computing. “On premise” computing reducing to “only required data/ computing to satisfy the limited operational time span, e.g.| 7days even shorter” everything else will be “off premise”. The landscapes will enable users to view content device independent and across the landscape, but drilling to a plant and process, collaborating across the virtual team from site to corporate to worldwide experts.

Certainly over the next 2 years the expansion in Invensys’s portfolio capability spanning local to enterprise landscape, and include both “on premise” and “off premise” as well as managed services as not a specialty but a natural way to grow value from your systems.