Digging for Treasure

Digging for Treasure
Alida Cameron
EPCOR
10065 Jasper Avenue
Edmonton, AB T5J 3B1 Canada

Abstract
GIS implementation projects are often embarked upon as a solution for many of the utility’s problems. Technology is able to automate and improve many processes, but without the proper planning and understanding of what the issues really are, a system can be put into place that does not solve even half of those same issues. In 1999, EPCOR Water Services began evaluating its current GIS and the new technology available. One of the findings was that the design of the current system was not able to meet everyone’s needs and that data quality controls were not as stringent as they should be. Solely implementing new GIS software would not solve these issues. A thorough investigation into the user requirements and the current business processes was needed. In the fall of 2000, the digging began. Information was documented in various forms, and through various venues. The information was overwhelming. A software tool was found that could aid in the organization and analysis of the information. What risks, problems, and issues were found during this process? Were process improvements possible? Was the information a treasure? These questions and more will be answered in this seminar.

Background
EPCOR Water Services is a business unit within the Canadian based utility EPCOR – also a supplier of power and gas. In 1999, EPCOR Water Services began evaluating its current GIS (1994 technology) as well as the new technology available. This evaluation was done through interviews with users, analysis of the current GIS data, and obtaining consultant information about new technology. It was discovered that not everyone’s needs were being met and that not all the data was structured in a way that other applications could use it. When looking at the results of the evaluation, it became evident that replacing the current GIS with a new system was not going to solve all the issues either. For Example, the data in the current GIS did not have the appropriate relationships to enable an automatic load into the Hydraulic Model. Instead, the Hydraulic Model update required weeks of data scrubbing. Also, many “work arounds” had been implemented to support the business (i.e. assets could not be recorded as they were in reality to accommodate a calculation or representation). Replacing the current GIS with a new GIS would only solve certain issues, such as allowing for a relational database. On its own, it would not be able to meet the business needs. It was decided that a more thorough investigation was needed to ensure that the new system would be able to meet both today’s needs and future needs. Hence, EPCOR Water Services embarked on a GIS 2 nd Generation Program.

GIS 2 nd Generation Program
With the advances that have been made in technology, it was decided that the GIS Program would need to take on more of an enterprise approach. The scope of the GIS Program became all data and business functions associated with EPCOR Water Services’ distribution assets. Therefore, it also included the modification or replacement of any information systems that contained distributed asset information.

The main objective of the GIS 2 nd Generation Program was and still is to provide simple access to information, and enterprise-wide access to data through viewing, query and analysis tools. This implies a common data structure and open architecture.

Getting Started – The Excavation
To begin the GIS 2 nd Generation Program, the first step was to determine the main business processes in Water Distribution. Each process was described in terms of its:

background - why and how it is done
findings - any information about improvements over the years or in answer to specific questions
questions and issues - any issues with the way things are done or questions about how things are done
people involved - a list of those who are involved in the process
action items - a list of any action items that could be done immediately to either answer a question or resolve an issue

The business processes were grouped mainly around capital programs or assets. In this way, the processes often ended up encompassing cross-functional areas. The first draft of these documents were completed by the project team and some subject area experts. As the business process descriptions were from only one or two individuals’ viewpoint, all the descriptions were sent out to the employees in Water Distribution. They were also posted on the Intranet and a common shared directory was set up in order to allow edits and comments. Many of the office staff responded and made corrections to the documents to help the project team understand the business processes. The response from the field staff was not as favorable. This was determined to be acceptable, as the next stage of the project would demand their involvement.

Through the documentation of the business process and issues, some action items or “quick wins” could be implemented. Certain modifications have been made to the current system to improve user access and searching capabilities. In addition, some data was identified as needing cleanup and what data cleanup could be done in the current system is underway.

The Digging
Once a basic understanding of the business was obtained, the details were ready to be flushed out. Before beginning to collect the details, the objectives and questions needed to be defined. The project team wanted to keep the big picture in scope and hence limited their analysis to any activity or piece of information that has to do with a distributed asset (pipes, valves…). The project team also had the following goals in mind:

document all business processes that use spatial data
create a data dictionary of the desired state
perform a CRUD analysis at the entity level
identify opportunities for optimizing existing processes
document required functionality for anything associated with distribution assets

More specifically questions like the following also needed to be answered:

What information is required to describe the distribution assets?
What business functions are associated with distribution assets?
What people are involved in the above mentioned business functions?
What information do they need to do the above mentioned business functions?
What existing computer systems support the business functions?
Etc.

The specific questions helped the team understand what pockets of information (people, information (data), business functions, existing computer systems, etc.,) needed to be gathered and what types of relationships between those pockets of information needed to be gathered (i.e. activities linked to people).

In order to collect the information, the project team initiated user workshops based on the business processes and invited cross-functional groups to attend. Twenty-seven workshops were completed to gather information on business functions, people, and information required by certain people to perform a certain task. The information gathered in the workshops was then placed in a database tool to help with organization and analysis.

Due to the cross-functional groups that attended the workshops, communication was improved. Not only did participants learn what others were doing, but also they began to understand why they were required to do certain things, i.e. data gathering. This improved communication continued into the workplace, and the different groups have begun to work together as a team.

Through the workshops, other “quick wins” were also found and implemented. These ranged from providing better data access to users to consolidating data sources.

The Dusting
Upon completion of the workshops, the post processing began. The information gathered in the workshops had to be consolidated and organized. Throughout the workshops, participants would use different terms for the same item and would describe their tasks in different ways. These all had to be consolidated so as to avoid duplicates. During this process some assumptions were made which will need to be validated with the original participants. Other information had to be gathered outside of the workshops. This included information such as existing computer systems, data sources, etc. Also, some of the information had to be created, i.e. automated functions, logical data model entities. These also will be validated with the users.

To speed up the data gathering during the workshops, the project team allowed for references to forms, drawings, and maps instead of the information directly. During the post-processing phase, interviews were held with users as to what the exact information on the form, drawing or map was being created, used, reviewed or maintained by the user. If this information had been gathered during the workshops, they would have likely taken a whole workday each, instead of the average 3-hour session.

At this time, the project team is still in the middle of this post processing. Some of the information still needs to be consolidated and organized. The next step is to validate the results with the users.

The Treasure
It is hard to convince upper management of the value of this activity, but you can end up with a treasure. The information gathered is enabling the project team to create an enterprise logical data model for EPCOR Water Services Distribution. It also will be used to develop the GIS Strategic Plan. The exact user needs and automated functionality gaps can now be evaluated to ensure that the highest priority items are implemented sooner. The information gathered can also be used as a planning tool for any information system or application implementation for Water Distribution. It will also be used and expanded on during the design phase. The physical data structure and the user interface screens will be designed based on this information.

The value of having this information collected in a database format and organized logically allows for quick searches. For example, if EPCOR wanted to know:

what activities are done by the Private Development Group
what forms or screens are currently being used to do that work
what information is needed to do that work

The database is able to display all those linkages at a simple click of a button, as shown in Figure 1: Private Development Activities.

Figure 1: Private Development Activities

Another example would be if EPCOR replaced their current GIS database:

what information is currently stored in the database
who would be affected
what business functions would be disrupted
what application interfaces would be affected

This information can also be easily displayed as shown in Figure 2: GIS DataSource.

Figure 2: GIS DataSource

The other advantage of having this information collected in a database and from a high level enterprise viewpoint, allows it to be filled in more completely from other sections, and non-distribution asset related activities and information could be added. This allows it to be used for many other types of projects, such as multi-skilling or succession-planning projects.

Conclusion
Accurate information is a necessity for improved decision making. In order to obtain accurate information, the data gathered needs to be answering clear and concise questions. The information needs to come from the users themselves and needs to be reviewed by them throughout the consolidation and preliminary and detailed design phases. It is important for as much effort as possible to be put into this so that you can end up with a treasure in the end.

Also, having a tool to help store all the information and relationships allows the information gathered to be more versatile. The information is readily available and answers to various questions can be easily obtained.