Water Industry and Data Analytics - Why Should We Take It On?

I have decided to write a number of posts and share with you the analysis done about how we should leverage the potential of data analytics as concrete and useful as we can.

A post, article or just a program on the TV warns me about the impact that “Data Analytics” or “Data Science” revolution will have on the way we solve our business problems. This happens so often. Every time more. The Water industry case is not going to be any different.

That’s why I have decided to write a number of posts into my blog, sharing with you the analysis done about how we should leverage that potential as concrete and useful, as we can.

In this one, the first, I would like to answer why we should do it, what are the reasons that should move us to it. In the second I will answer what we should do and explore what’s the expected return from Analytics investment. And in the third one I will detail how we should do it, specifying where we should apply it to obtain the maximum social, environmental and economic benefit in the least possible time. To do this, I start with a simple question:

If I had to apply the current benefits 'Data Analytics' is offering into Water Cycle, what would be my priorities?

This simple picture is my answer

These are, among many other possible options, the points where I think Data Analytics could provide the max value in less time. Some of these have been analyzed in the past and are in a relatively “mature” state. These are very common problems. However, the paradigm change opened by current analytical and technological possibilities (IoT, Big Data …), makes it mandatory to rethink many of our old fashion basic hypotheses.

Thus, my response and the value estimation done, is the result of a deep reflection that has taken into account several aspects such as being able to guarantee the maximum water service guarantee, maximize customer satisfaction, minimize environmental affection, maintain a sustainable operation over time, ensure economic and capital planning viability and take advantage of technology adoption opportunity cost. In short, I’ve taken into account the  future major challenges for Water Industry

The solving process has been very interesting. That’s true. I consulted multiple sources of information along the journey taking as a basis simple questions and getting something valuable at the end: knowledge and a response I would like share with you.

But first of all I need to express an opinion that surely isn’t going to be to reader’s liking. Unfortunately, the Water industry is not a “cool” industry. We can assure without fear of being wrong that our industry is not at the forefront of  technology adoption .

And Data Analytics is the trendiest and coolest thing right now. Rapidly I can name few industries with a firm commitment and investment in Data revolution and where data analytics projects are being intensively tackled. In contrast to these, I can think of few industries as critical to humans as water.

This fact makes that a career into the Water industry isn’t the top priority for data professionals. There’re many more attractive sectors doing much more advanced things nowadays. We don’t have too much options. It’s time to start believing in Data Analytics and invest decisively in this direction, otherwise we’ll be out of order (or data) soon.

And why should the water industry invest in this field?

That was my first question. The answer is simple: because we have to analyze things to understand them properly and finally be able to make decisions according to evidence, not with assumptions. This is the radical mentality change: from what we assume to what we can demonstrate.

There are several ways to approach analytically our daily problems resolution, from the simplest to the most complicated. So we apply the analytic to:

• Analyze in the most traditional way: obtaining a global vision of an entity or a business process, based on the historical indicators visualization, It’s about trends, alarms and thresholds. It is what we know as  DESCRIPTIVE ANALYSIS .
• Compare between elements or options taking as a basis their Analysis model indicators and to segment or classify these elements based on patterns or profiles definition, obtained through different analysis criteria. It’s what we know as  DIAGNOSTIC ANALYSIS
• Forecast events that are going to happen based on the knowledge obtained from previous analysis. This is what we know as  PREDICTIVE ANALYSIS .
• Simulate and recommend ways to proceed or act based on the knowledge gained and to optimize business processes operation or efficiency. This is what we know as  PRESCRIPTIVE ANALYSIS .

The adoption of these stages imply an organizational maturity. The added value comes from the knowledge they acquire progressively during the path.

The most advanced are already at the point to tackle prescriptive analysis projects, that is, to automate the decision process to the maximum according to the knowledge acquired by the organization, both internally and externally.

It’s time to move from predictive to prescriptive, from predictions (a specific event probability), forecasts (predicting events in time) and simulations (predicting multiple events with an emphasis on uncertainty) to rules (to get a predefined framework to decide between alternatives) and optimization (an evaluation of interdependent options in a results oriented way based on their limitations).

Water Industry, except for honorable exceptions like our fantastic hydraulic models, has been involved for many years in the “description” phase, always asking about what has happened. It’s perhaps the solid foundation on which the data science has to be built, so this phase is already a great challenge for Water world. This industry has long been embraced tools oriented to go deeper into the descriptive analytic and intuit the diagnostic part. But there’s still a long way to go.

So the question is  what’s the path you need to take to get the most out of your data?  In the next post we’ll jump on this exciting topic.

See more on: dcar Absolutamente