April 2007, Vol. 19, No.4
A Balanced Approach to Automating Small Systems
James C. Pyne
The wastewater industry has undergone dramatic changes in the last 30 years. Digital technology has entered every aspect of wastewater operations and maintenance and increased our productivity at every level. Some tasks that once took hours and required miles of walking on large treatment plant sites — or driving throughout distributed wastewater systems — have been reduced to minutes. Automated systems perform many routine procedures, enabling utility personnel to concentrate on more creative endeavors. However, all this convenience comes at price, which can vary depending on the situation.
The Hampton Roads Sanitary District (Virginia Beach, Va.), is expanding a wastewater treatment system for a small population center in a relatively rural region. Let me share with you some of the issues we’re considering.
The first question we asked was, “What’s the objective?” or in management language, “What return do we expect from our investment?” We expect that an automated system will cost less to operate and will offer greater reliability than a comparable system without this technology. Specifically, we expect fewer operator hours onsite and optimized usage of power and chemicals — the big-ticket items in our operations budget. We also expect operating costs will decrease with fewer “alarm call” trips to the facility. Remote distributed systems such as ours may require at least an hour of round-trip travel time to respond to an alarm call.
Small systems have much lower revenue and operations budgets than larger ones. Therefore, while automation may be less costly to implement for smaller systems, it also may yield less cost savings.
This is complicated by the fact that regulatory agencies usually require a minimum level of staffing, both in terms of hours onsite and personnel qualifications. Some of these requirements may be negotiable, depending on the level of automation and monitoring included in the facility. Therefore, as we increase the level of automation and monitoring, we should realize a proportional decrease in the manpower requirements, to a point. There will be a minimum manpower requirement that is either prescribed by the regulatory agencies or, if “unattended operation” is a possibility, needed to perform maintenance on the mechanical components of the facility.
The goal is to optimize the level of automation and monitoring in distributed systems to make the best use of the available technological, human, and financial resources. Not all of the instrumentation technology available for distributed systems will result in a net increase in reliability or reduction in cost. Below are some issues to consider when deciding on the optimal level of automation and monitoring.
As automation and monitoring increase, so does the need for instrumentation technicians. In our case, instrumentation technicians are paid at a higher grade than plant operators, so a decrease in operator effort is required to offset the increase in technician effort. The goal is to optimize the costs among plant operator effort, instrumentation technician effort, and the equivalent annual cost of installing the various levels of automation and monitoring.
Remote monitoring has its own set of considerations. The level of monitoring required may depend on the complexity of the facility. Monitoring ranges from the simple transmission of basic alarms to complete control via advanced supervisory control and data acquisition (SCADA) systems. In our case, moving to an advanced SCADA system capable of more than basic alarm functions and manipulation of rudimentary controls would require replacement of our existing system, a costly effort that our budget would not cover. Therefore, we must compromise and choose monitoring and alarm functions that suit the bandwidth of our existing SCADA system.
It is important to consider the proper operator response to the various readings and alarms transmitted over the system. Some alarms may require only acknowledgment, such as a notification of “commercial power failure” quickly followed by “generator start.” In that case, the operator only needs to be aware of the condition and be alert to further alarms, such as “generator fail,” that then would require the operator to respond to the facility and take action. If the same response is required for a number of different alarm conditions, then it may be possible to combine several individual alarm conditions into a common “trouble” alarm that alerts the operator to respond to the facility.
Because we have the capability, there is a tendency to want to know everything that is going on at a facility all the time. There comes a point, however, when we reach information overload, or as we say, TMI — too much information. TMI raises two concerns. First, gathering the information through technology has a significant cost, as does the effort it takes to review it. Collecting more information than can be analyzed and acted upon in a timely manner is a waste of resources. The second concern is the nuisance factor. Alarms that don’t require immediate action quickly can become a nuisance that operators start to ignore. Unfortunately, operators can ignore the important alarms along with the unimportant ones. At night, when numerous nuisance alarms have roused operators from a sound sleep, pagers or cell phones often get turned off “accidentally.”
We have made great strides in improving the efficiency and reliability of our utility systems through the use of digital technologies. These information systems have become an essential operation and management tool. However, like any tool, they must be applied properly to allow facilities to operate at peak effectiveness. The levels of automation in distributed systems may be different than for large central systems. And reducing accidental alarm calls through automation has the potential to yield significant savings. However, these savings may be overwhelmed by implementation costs and the ongoing costs of maintaining the system. Different organizations are capable of supporting different levels of automation and monitoring, depending on their size, organizational structure, and financial resources. When considering what level of automation and monitoring to invest in, it is important for organizations to consider these factors, as well as their operational needs
We must be aware that essential parts of our information management systems include the human elements that utilize the information. It is through joint optimization of both the technological and human parts of the system that we can maximize returns in our investment in both of them.
James C. Pyne is chief of the Small Communities Division at the Hampton Roads Sanitation District (Virginia Beach, Va.).