Migrating Legacy GIS; an Evolutionary Approach
William Shaw Manager GIS Mapping We Energies 333 W. Everett St. Milwaukee, WI 53290-0001 Telephone: 414/221-2379 Fax: 414/221-4543 Email: William.Shaw@We-Energies.com Abstract: Like many of today’s Electric and Gas Utilities, We Energies (formally known as Wisconsin Electric Wisconsin Gas) is continuously confronted with the need to integrate, merge and evolve its core legacy systems to meet the demands of new business requirements and the challenges of effective implementation of newer technology. To this end we have chosen an evolutionary approach to migrating three legacy Electric GIS systems into a single common source for all Electric GIS data. The legacy systems that are being evolved include a working GIS maintenance application and data reporting system, a traditional CAD Electric Dispatch mapping system and a recently implemented Outage and Distribution Management System. The project team at We Energies has come to appreciate two tenets of today’s technology as it relates to effective business implementation and those are that ‘it’s a mixed bag at best’ and ‘that bigger is not always better’. With an eye toward limiting our risk exposure we have managed to overcome numerous technical hurdles while at the same time leveraging old and new technology to arrive at a best fit for our Electric GIS applications. Project background Beginning in November 2000, We Energies began to implement a strategy to eliminate the maintenance of multiple electric facility record keeping systems. The implementation of this strategy prepares the way for establishing a single point of maintenance and control for our critical spatial and asset inventory information. The strategy is targeted at achieving a principle objective: to provide effective and reliable support of field mapping, outage management and dispatch operations. The evolution of the existing legacy mapping systems at We Energies is probably not unlike the GIS development patterns experienced by many in our industry. GIS is first brought in because GIS technology is unparalleled in its ability to integrate almost any aspect of daily operations. Unfortunately to get GIS ‘right’ there needs to be an understanding that it is also a terrifically well-suited disintegrator. The good and the bad tend to work against each other and until an organization discovers this and acts to effectively coordinate its’ spatial information the more issues that can arise to stand in the way of effective progress toward it’s enterprise goals. Observations The latest and Greatest Competition and opportunity drive the GIS industry to continuously improve and upgrade their approach to GIS solutions. This is a good thing overall and it’s ultimately what sets the various vendors apart from their competition. Some of these approaches are possibly better than others but any mainstream approach is by and large capable of being molded to meet the majority of requirements. This leads to the first fundamental tenet of the evolutionary process and this is ‘it’s a mixed bag at best’ which means that no one vendor is likely to have the perfect solution. In other words, there’s a good possibility that a multi-vendor solution is in your future. Another tenet of the evolution process to keep in mind ‘is that bigger is not always better’ and its corollary ‘sometimes it’s just impossible’. It’s been our experience that as the size of the evolved application grows, the capability of the solutions are exponentially more complex and difficult to implement. Unless you are prepared for this you could find yourself further out on a limb when you have to deal with hardware/software performance, standardized products, capability limits and the elapsed time to implement. This in turn eats into your Return-On-Investment to point that it becomes pointless. So the advice is to take significant but small steps toward your objectives. Our experience suggests that the industry as a whole is getting better and is more apt to have answers today then 5 years ago but unfortunately a few of us are still dealing with the systems that were installed 5 or more years ago. Finally, the GIS industry is driven toward newer technology and there is a dynamic between new technology sales and supporting the old worn out stuff you bought years ago. Did I say that in our case that it took us this long to get it to work? Conversely, what’s new in GIS falls into the category of process improvement, which apparently some feel has a lifecycle of two years. Do the math. With evolving systems taking more than 5 years to fully implement this could mean that you would be somewhere in your third process improvement cycle before you had finished implementing your first. Obviously there is something more fundamental to this than simply implementing GIS software. We’ve discovered that the fundamentals lie in process control and data management. No matter what your GIS software supports, if you can continue to perform these two operations within acceptable tolerances you will be prepared for the next wave or at least limit your risk and/or minimize your recovery time. Legacy Musings It happens to everyone. Sooner or later the weird package you bought (think of Synercom or Genamap) and the doubly weird stuff you did to it to make it work is just not getting the job done anymore. So what to do? How do you mold whatever you did before into whatever you’re bound to do in the future? The answer is it depends and that answer is dependent on whether you believed any of the hyperbole that was alluded to earlier or if you really do intend to address the fundamental elements of GIS evolution. The caution here is to keep in mind that (write this down) ‘it’s the data stupid’. It’s also the process that delivers that data to you and you need to control both. If you concentrate on the data and manage the workflow to maintain the data integrity you simply cannot go wrong. In other words, whatever is left over after all is said and done is not the GIS software, its’ the data and the infrastructure that supports the data. It’s the way your crews process the work reports they complete, the crew chiefs willingness to report process changes up and down the line, the trouble crew’s attitude about moving something as insignificant as a secondary split and the process of reporting the change. It’s all the plant accounting, auditors and recording specialists all along the way and how concerned about the value and worth of their part of the process. The data, when you thoughtfully consider its importance, needs to be alive in these environments and it must be cultivated and nurtured to keep it healthy under the worst of conditions. We Energies has begun a course that we expect will evolve to successful GIS solutions. Although we realize a need for new technological solutions we are consciously looking to simpler solutions that we can test and evolve within our enterprise suite of technology and integrated process. At the same time we are looking to see how robust and extensible these new platforms are and how they can be extended to do more productive things in ways that were not imagined possible just a few years ago. The evolutionary process Establishing connections between enterprise solutions Taking meaningful steps toward integrating enterprise solutions requires a determined approach to selecting and implementing software and process over many years. Taking the long view is required because if you don’t it will become nearly impossible to get started. While company mergers are probably today’s strongest catalyst there are other situations that could also become reason to re-examine process and technology. Basically any major technology or process change that makes a significant change to the current way of doing business is a candidate for opening the door to developing incremental improvements toward enterprise solutions. In our case the merger of Wisconsin Electric/Wisconsin Gas set the stage for adopting GIS system strategies that are moving the two companies toward future enterprise solutions. These strategies focus on building synergies between the two established companies and leveraging these toward our common goals. An example of this is the development of a common viewer technology for Gas, Electric and Water. This is being rolled out in 1st Qtr 2003 and was accomplished at a fraction of the cost of adopting a single GIS platform technology. A basic understanding of who would be impacted and where we could establish the largest benefit in the quickest timeframe lead us to this solution. Developing a culture within the organization that supports an enterprise view is also key to ensuring strong connections among enterprise solutions. There’s possibly nothing more difficult to overcome than the effects of frustration and inefficiency brought about by obviously disconnected systems and procedures. To overcome this there needs to be a concerted effort toward building bridges between systems so that it is apparent that every system that is deployed is focused on the elimination of these types of problems. In addition to eliminating inefficiencies, it must also minimize the impact of any new problems of it’s own. This is best accomplished through deployment of systems that thoroughly support new work procedures that drive toward solving real work issues as much as improving back office reporting and immediate feedback. Determining data and process ownership among system components Interesting things happen when there are clearly defined owners of the various system components. Accountability and performance measurement becomes a norm, which in turn builds pride and teamwork resulting in a job well done. It’s so important that to be successful there needs to be a clear and direct understanding of each employee’s role in the process. This may appear to be overstating the obvious but without it any enterprise implementation is doomed. At We Energies it was clear that we needed to establish this linkage between process and data elements. The 3rd Qtr 2003 implementation of Work Management will provide the mechanism for efficient process controls that are lacking in the current environment. That mechanism relies on consistent and developed assignment of ownership up and down the line. The GIS applications that are coming on line during the same period will benefit tremendously from the Work Management initiative. We will for the first time be able to see all work coming through a common system and reporting mechanism. An example of this is that our work package contents including original sketch and as-built products, will be immediately communicated to the field through our common viewer application. Integrating old systems with new ones while keeping the infrastructure afloat Evolving enterprise systems requires that you develop good working relationships with the mainstream vendor community and establish strong partnerships with a solid, experienced consultant team or systems integrator. The vendor and vendor support community is committed to seeing these projects through and can be relied on to assist throughout the process. They can help you weigh the risk of the incremental adaptation of new technology versus the risk of implementing the same technology more quickly but at a much higher cost. They can also ensure that the technology supports the current business operating culture. The risk of not quite fitting into your culture may be high and therefore the new technology is not a good choice, at least at this time or without limited introduction. An alternative, in many cases is to grow your current applications in the direction you intended with only implementing selected parts of new technology. The real challenge is to make adjustments in the existing systems that align themselves with the outcomes that are desired. It is evident that the key to enterprise solutions rest in the data, process and people that support the organization. Technology is an enabler to a certain extent but with regard to the process underway needs to address five primary objectives:
This project prepares the company to address the current state of maintaining multiple sets of records of our primary electric distribution network. There are multiple inconsistent record sets with no single complete and accurate set. Eliminating inconsistencies between the systems will improve accuracy of operation, as well as reduce operating costs of redundant entry of data. History of related systems at We Energies:
1990 CableCad (PC based) introduced to enhance DFIS graphics and editing 1992 GIS (ESRI ArcInfo) chosen to replace DFIS (IBM abandons product) 1995 Conversion from DFIS to ARCInfo GIS begins 1996 CADOPS (ABB) selected to provide DMS/OMS solution. Begin data extraction from GIS to populate CADOPS database 1999 Conversion from DFIS to ArcInfo GIS complete (exception Dispatch Redline) 2000 Dispatch Redlines (CableCAd), DMS/OMS (ABB CADOPS), GIS (ArcInfo) 2001 Population of CADOPS database for all areas complete 2002 Continuing Maintenance of three existing systems Design and implement an interface process between the current GIS and CADOPS computer systems to allow for data maintenance operations performed in our current GIS to be transferred to CADOPS. This includes the preparation and conversion of GIS data to be compatible with the Dispatch CableCad and CADOPS systems, as well as the development of necessary tools to perform this conversion.
In conclusion the evolution of legacy systems across the enterprise today is made easier due to the advantage of having better technology and more experienced vendor and integration support. Software solutions are significant but may not be the most important part of implementing enterprise solutions. The majority of issues are found in process and data management and these need to be addressed somewhere in your plans if you are to be successful. Evolutionary steps in your implementation plan provide an opportunity for an incremental approach which allows you to address data and process problems over a longer period of time. | ||
|
|