GE Intelligent Platforms is a leading global provider of software, hardware, services, and expertise in automation and embedded computing for different industries, including water and wastewater. The company's Global Water Industry Manager, Alan Hinchman, spoke to Anoop K Menon on new offerings related to the water industry, especially in the Proficy software platform and shared his insights into developments from the customer side. He was accompanied by the company's Marketing Manager for Middle East & Africa, Joseph Davis.
Could you brief us on the key aspects of GE Intelligent Platforms' focus on the water industry?
GE Intelligent Platforms is the controls and interface business of GE. We work with customers in the water industry to build out their control systems, and work with them to better manage their operations, and help them provide all that information back to their business and operators. A key area we focus on is the conveyance of water, how the water is being pumped or providing information around demand and usage, and pump percents to enable greater efficiency or looking at the use of distribution system for inventory management or even understanding the quality of water in the system.
The treatment side too is an important area for us. We have developed a new DCS-style controller called Proficy Process Systems, which allows operators to distribute the control and have a single database of information. A typical controller reads all the code and eventually when it gets to that function, it executes that function and then continues down, and reads through that. You may have a few milliseconds of variation between the times it scans through, if it is waiting for some other piece of equipment or message. However, our controller stops what it is doing and goes back on its timed approach. In fact, our first customer in the US reported a 15% reduction in chemical usage, the reason being that it allowed them to give a tighter tube under the chemical management system. By closing the window, you have less overage and less underage, which causes overage. By reducing your control window, you can flatten out the chemical usage.
Another aspect of treatment is protection of membranes from clogging, extending their life and extending the time between backwashes. Recently, we acquired the technology assets of CSense Systems of South Africa. They have a predictive software product, which can look at a series of events, understand if they are relative to each other and if so, look at how the information can be leveraged to predict what is going to happen. From that we build a model, which says, that when the water is of this condition, you have this length of membrane time. It allows you to predict the management. An interesting case is a customer, who is now using this to mix feeds. Their two surface water feeds had two different profiles, so they decided to check at the source. They found that one at low tide would pick up a lot of mud and biological material from the water, while the other one would have a different profile. So they decided to blend the two profiles to get in a better mix that allows them to execute longer backwash times.
What's the deal on the operations side?
One of the things we have come across globally is that operator capability is actually diminishing. Also, we are seeing a significant drop in average tenures. Unlike in the past, operators today work for only short periods in the water industry before moving on to lucrative offers in fields like oil refining, where their skill sets are in high demand. This problem of high turnover rate is being exacerbated in the US and Western Europe by the retirement of existing operators.
At the same time, more and more is being required of today's operators, whether it is monitoring the security or the water quality to ensure that it is line with regulations or specifications. In many countries, they are licensed and measured on a number of these things. Moreover, the amount of data they have to digest causes a 'paralysis' of sorts. More often than not, they end up restricting themselves to specific things that landed them in trouble before. For example, if it was low chlorine residual that caused trouble in the past, they would focus more on making sure there is plenty of chlorine. This leads to sub-optimisation, so that no matter how good your control system is, there is no way of getting back to the optimal level.
Last year, we developed a product called Proficy Workflow, which provides operators with what they need to do on a daily basis. It has something we call electronic Standard Operating Procedures (SOP). For example, if you need your field resources to go and take a sample, how do you know they will take the sample correctly? What Workflow allows you to do is walk them through the task, thereby building a chain of custody and ensuring proper completion of all the work steps.
A nice thing is that Workflow has no database of its own. Rather, it writes to everyone else's database, so you can see the connections. We can take a Laboratory Information Management System (LIMS) or maintenance system and write that information back. If there is a valve or reading that needs to be reported, it writes it back to that piece of software, which means you don't have a stack of different programmes that have to be combined.
Workflow also has an Alarm Management capability, which allows customers to automatically respond to alarms and exceptions from multiple systems. One of our customers, during installation, added plant start up and emergency plant shut down, in addition to standard parameters like chlorine residuals and turbidity. This summer, a large power outage in the area forced them to shut down. Thanks to Workflow, they had everything they needed to start the plant, saving them the over time of bringing in a super experienced operator to re-start the plant. When they began to aggregate that over last year, which was their first full year of operation, they found that it reduced one operator. The fact that a piece of software gives you an ROI of eight months against the cost you have incurred on it is a phenomenal thing in the software world.
Using Workflow, a customer in Miami combined their power, water and reclaimed wastewater departments under a single call-in number. Now when someone calls in with a complaint, Workflow walks the person at the other end through the steps that a phone operator should take, in terms of the questions that needs to be asked, and also understand how the complaint is to be routed. That customer is seeing a much faster response to water leaks because they can process calls in a much more efficient manner.
In South Africa, for a rural customer, we went a bit further by adding SMS capability to a Workflow implementation. This enabled them to use Workflow to strengthen customer engagement. For example, based on a criterion, say, a pump going down, which could lead to water outage in a certain area, Workflow can guide operators to alerting the affected residents of that area. In this case, the software will ask the operators if they want to message the residents. If the operator says yes, the software will send out an SMS to the affected residents' cell phones informing them of the outage and its possible duration and also advise in terms of collecting and storing water for the duration of outage. After the issue is resolved, the software will ask the operators if they want to message the residents that the water supply is restored. If they say yes, it will send out a second SMS. Interestingly, their revenue increased three times once they discovered postal service issues in these rural areas had resulted in many customers not getting their bills.
Joseph Davis: You are able to correlate the real time information to an isolated segment and inform the users within that segment. In other words, it doesn't become a spam going to everybody, affected and unaffected alike.
In a recent case, we partnered with a long-term customer, Waterford Township Department of Public Works, Michigan to tie their information system and GIS system into their control system, so that the area where alarm originated can be pinpointed on a map as well.
Thus, from an operations perspective, the main concerns addressed are how to reduce the risk, how to ensure there is no over-time or avoid having to call someone in, which could increase down time. The other aspect is that when things go horribly wrong, a key concern would be - how are you as a utility making sure all the rules are being followed. Ultimately, through Workflow, you can have an operator who becomes more efficient and as an extension of that, it also allows you take a look at your automation investment.
Are you saying that we can save on automation expenses through the deployment of Workflow?
A number of things throughout the water system haven't been automated because the cost versus benefit equation didn't work out. In many of these cases, cost of automation is probably much more than the data itself. But when you look at remote sites where you will need this information, by using Workflow, you could avoid incurring the costs of putting in controllers. You can have a system that your operators can carry on a tablet PC, and allows them to see what is going on and interact with the software.
We are also trying to push out operational information to the field through Workflow. Some of our customers are using it for meter readers, a few others are using it for their LIMS. In the latter example, if they are going out to do testing, they can do that in the vehicle, and enter the data there itself instead of putting all the samples in the vehicle, driving back and typing it in. Again, if you decide not to automate, for example, the measuring of level in a lift station, and instead have somebody checking it on a route, you will end up setting up lots of routes and all that the field force will be doing is checking this flow meter or that water meter. What workflow allows you to do is to control the routes and if some of these don't make sense, you can capture that information in a centralised database instead of a piece of paper sitting in the back of truck. You can start to leverage that information just like it is automated because you have got an auditable process.
Workflow is also facilitating the creation of a mobile workforce, a very interesting concept for the water industry, which has assets spread literally over hundreds of square kilometres. With Workflow, we are trying to change the way that operators are working with all that.
Joseph Davis: Also, industries where we have deployed this software have not really lost people. They have simply moved them to higher value activities. I have seen people who were data collectors move into continuous improvement activities. They are now re-looking, based on their expertise in operations, how to improve the process, how to make it lean, how to eliminate unnecessary steps because Workflow becomes an enabling tool for them.
With Workflow guiding operators at every step, will they have the liberty or flexibility to put their practical experiences to use?
The way most of our customers prefer to implement the software is to leave their SCADA system on its own servers, on its own screen, and have Workflow running on a second screen. Workflow can embed on itself the start and stop buttons of different aspects of the control system. If you have an operator who wants to work through Workflow, he can have these buttons on his screen. However, if he encounters another set of conditions, where his experience tells him that there is another parameter that needs to be changed, he can still do that. However, that's up to the individual utility to understand, but they can enable that with a password. A senior operator could bypass Workflow, but if it is a fairly junior operator, they could insist that he sticks to Workflow. At the same time, one should understand that Workflow is more like an assist rather than an Autopilot in a plane.
Can companies use Workflow to capture the knowledge base of workers?
Joseph Davis: In the case of a well-defined procedure, the operator just follows the procedure and in some cases, the logic may branch out based on his input or where he needs to confirm certain things. The other implementation of the software, which we call Troubleshooting Trees, is akin to a Decision Support System (DSS). It keeps learning and gives suggestions, but the operator still needs to act because the correlation is not clearly defined. It is more of model, which keeps on evolving based on the variables.
What is the scope for a product like Workflow in this region?
Joseph Davis: The biggest opportunity, in terms of quantity, lies in the controlling of production because without production, there is nothing to optimise. There is the full value chain, from pumping in the sea water, the conveyance mechanism, the whole treatment and the point of use, where it becomes wastewater, followed by collection, treatment, disposal or reuse as the case may be. Maintaining the throughput in a reliable way is the foundation business.
But with the growth of urbanisation, the scale of investment and also equipment is getting bigger. Assets are very critical because they are of bigger scale to get the value of scale. Here, we see great opportunity in terms of people using additional information, additional layers to protect as well as to get more out of these assets. So we see the need to know, the need to connect and correlate growing because people are looking at things, not just from an equipment point of view but in terms of the interdependence of all this within the context of the enterprise. For example, people are interested in knowing what will happen if the demand for water goes down, how this would impact availability, when will this effect be seen in treated water, what are the lags in the system, and so on.
People layer that with the whole sustainability matrix in terms of how to balance demand and response or how to bring in the energy angle into this discussion and in other ways.
Also, they really want to get visibility into their operations. From the maintenance point of view, with capex reducing and opex increasing, they are interested in knowing the availability of equipment and the total maintenance that has been carried out.
The next level of this is knowledge services related to big capital equipment, the predictive modelling side, which helps to tell in advance whether any equipment is showing signs of failure, so that corrective action can be taken in advance.
The way I see it, we would have a pyramid, where the base would be standard production; in-between, we have the visibility layer as people want to be sure of returns from their investments, see which area they should invest in, and the incremental benefits. In the third layer or top layer, with limited users, we would see knowledge-based services, which is more of services and less of software.
© H2O 2011
