GISdevelopment.net ---> GITA 2001 ---> How They Did It - And What's Next

From the GIS-project to everyday life

Torunn I. Carlsson
Oslo Water and Sewage Authority
Postbox 4704, Sofienberg, N-0506 Oslo, Norway


Introduction
Today the Oslo water and sewage authority (OWSA) has a enterprise-wide information system, VIS. With this system we are now able to provide accurate, detailed information about the city.s water and sewage systems. The thought of VIS started in 1991, and the first modules were up and running in 1993. This paper will tell a little about the setting of the project, i.e. the organization around it, and also about the project itself, and what has happened after the project ended, as the operation and maintenance part of the system life began. Finally there is a little about the future; what will come next for VIS?

Oslo Water and Sewage Authority (OWSA)
OWSA has the following vision for its business: "OWSA will supply clean water for all purposes"

This will be achieve through aiming at these business goals: "Customers and users should all be satisfied with OWSA.s products and services, including their accessibility, quality and price."

OWSA is an agency for the city of Oslo city council. It is fully self financed. Our main income is water and wastewater fees. We supply approx. 520 000 inhabitants and all the businesses in the city with water and wastewater removal.

Digital history of OWSA
OWSA started getting prepared for the digital age in 1972. At this point the organization started to register coordinates for all manholes from aerial photographs, and the .Water and sewer register. (VAR) started in 1973. In the period from 1982.83 the first interactive graphic map system was introduced (Mivareg). This was used to register all pipes and manholes from VAR. By 1991 100 % of the utility maps were digital, this also included some basic attribute data. In 1991 the division's, and the different specialist fields in the technical divisions had some specialized systems. There was little or no sharing of data, and the organization did not yet have one PC or workstation pr. employee.

The VIS project setting
The VIS project was initiated at a time when there had been a major reorganization in the agency. The agency was trying to change from being a supplier of water and wastewater removal, into becoming a customer oriented service partner for the city, in all aspects concerning our business. The executive committee was very keen on utilizing new technology to achieve both better services and also higher efficiency. There was general agreement that a good strategy was to connect all the dataentities using geography as the common denominator.

The organization in 1991
OWSA was organized in four operational divisions and one support division.


Figure 1 OWSA organization in 1990/1991


We had a total of approx. 530 employees. The Customer service (CS) division was where the VIS project was located. This was done to include the customer needs as well as our internal needs. The mapping section is part of the CS division. The mapping section and the staff of the CS division were the .core. workers into the VIS project.

Keywords for the early project
The goals for the project was written as follows in 1991:
"The system shall support and aid OWSA in reaching its strategic and business goals."

"The system shall provide all employees with"
  • cost-efficient computer aid
  • Correct and updated information
  • Simplification of operations
  • Reduction in data entry/registration work.
The system should be able to handle at least the following amount of data :
  • 1512 km Waterpipes, 666 km Sanitary sewers, 816 km Combined sewers, 572 km Storm drains
  • 45 000 Manholes, 105 pump stations,
  • 580 sheets Technical maps (Basemaps,(1:500,1:1000), Economic maps (1:5000), Overview maps),
  • 60 000 Technical installation drawings,
  • 1 mill Installation documents,
  • 30 000 Drawings (OWSA installations)
The VIS system goals
The system was to be based on the key thought "Data is the fuel in the information system", i.e. the solution should be based on one integrated data model.

The idea was to have an integrated system where all our data was stored in commercial databases, with easy communication to other databases maintained by different public agencies. Both databases, software and hardware should be " industry standard", and should comply with all necessary standards. The system should help support our business.

Some of the functions that were part of the initial project are shown in figure 3 below. In addition to these functions the bidders were told that the functions of some of the key systems already in use where part of the user requirements. One interesting requirement was a system to support scanning and later use of approx. 1 million documents about the 70000 properties in the city.

The VIS project
The VIS project was started by appointing the projects steering committee, and project leader.

Project organization
The steering committee for the project was identical to the agency's executive committee. Then there was a project owner, the director of the customer service division, and a project leader. (See figure 2.)


Figure 2 The VIS project organization


The VIS project was started with conducting a limited data-/informationflow analysis and an overall data model was made. The early project phase also included user requirement survey for the complete enterprise information system.

The VIS project sent out its invitation to tender in early 1992. The contract was signed in May 1992, with a GIS system supplier, to develop the VIS system based on standard software.

The project was based on heavy user involvement. The purpose of involving many users was to create an ownership to the solutions by their future users. There were numerous workgroups in addition to the more formal groups listed above. In 1995 there were about 50 different people involved in different subprojects. Considering that the planned total number of users for "VIS -first stage" was approx. 150, one out of 3 users were involved in the project.

The VIS system sketch below shows the 4 modules planned: Map-, Customer-, Water & Sewer- and "Office"- module. (By "Office" - module it is here meant a system for follow up on all correspondence and other inquiries to OWSA, document handling and archiving. It also includes document- and "work"-flow, but not word processing programs etc.)

The functions are grouped around the databases as shown in figure 3.


Figure 3 "VIS-system sketch" - main functional groups
for VIS grouped into the four application areas.


The project period was supposed to end with a full installation midyear 1994. Unfortunately the project ran into technical problems. The problem was solved by replacing the main supplier. The parts that functioned satisfactory were kept. By 1995 the office module and the module for scanned drawings and documents were established. There was still a lot to do on the map and water & sewage modules. By having had our employees involved in the whole process, most of the work conducted was still usable, and by December 1996 the .Waterworks Information system.(VIS) was up and running. The system included the data as shown in figure 4.


Figure 4 Data input to VIS in 1995


The main goals of the project were met. With the data input as shown above and VIS software, OWSA now had the tools to improve customer service. All spatial information about pipelines are integrated with the customer service module. We had analysis and decision-making tools for pipeline network, archive search tools and extensive map production tools.

Project challenges
The project had a serious setback when the supplier could not provide a good enough technical solution. (The response was so slow it was not even possible to test the applications) As an afterthought OWSA were probably a little a head of the technology in our tender. But technology caught up with us, and a new supplier of technology was found. Due to the heavy involvement of people in the organization this setback caused very little time delay in the project.

The change of supplier also caused some secondary trouble. Some of the technical personnel felt a strong ownership to the systems they had worked with for the last couple of years. This caused some turnover in a critical phase of the project.

The project was organized in the customer service division, which is not a very .technical. division. It was a challenge to sell the VIS solution to some of the specialist groups. Some of them also had to give up their .own. specialized system, for the benefit of data- and system sharing, and in some instances VIS gave a less perfect solution for each user. This challenge is to some extent still there.

Another challenge in OWSA was our high degree of computer .illiterate. employees when the project started. This challenge was met by the policy that everybody should receive sufficient computer training to do his job and participate in the project if needed. Employees were given the opportunity not only to take short courses, but also more extensive classes given at universities were paid for, and could be taken during work hours. This seems like a successful path to have followed so far. The people that has received this .on the job. education presents a very stabile workforce, with the knowledge we need. They have a willingness to keep on learning, as this has been part of their .job. during several years.

Transition between the project and running the system
The different modules of VIS were installed and ready to go by early 1996. The massive datainput and scanning were also almost completed by 1996. The modules for "office", scanned installation documents (Customer), maps and W&S was set into daily work, and the number of users in the organization was as follows; 125 users of "Office", 76 users of installation (i.e. customer), 89 desktop map users and 88 users of W&S module.

The list of applications available under each module now looks like this:


Figure 5. Functions in each VIS module en early 1996


The first year of operation (1996) was used for massive introduction activities. People were very enthusiastic, and there were a lot of new tools to learn. The mapping section, where the project organization was, had extra people in this period. After 1996, our number of employees has decreased, as the massive datainput, and training activities has finished. We now see that OWSA's need for a mapping section has changed from a datainput section to a GIS resource pool where they can get help with their specific GIS analysis or datapresentation needs. At present we are still having problems meeting the new needs in the organization. This is mostly due to lack of people, and to some extent lack of knowledge amongst the people that we do have.

Challenges in the transition period
Ability to fulfill new user needs and requirements
As stated above, we at present have very little capacity to fulfill new needs in the GIS area. We are working on getting an acceptance for the fact that we need to grow a little in the GIS field, even if there is a general cut back in the work force.

Keeping keypersonnel
We have been able to keep most of our key personnel even after the "fun" project period was over. This I think is primarily due to the fact that we can offer interesting jobs, and also continued education, if wanted. We can not compete salary wise with private companies.

Project organization for new subprojects
Another challenge has been to find a format for the (sub-) projects we have done since 1996. We have small projects all the time and the problems arise mostly when modules interact, or need add on's in the data model. We have not found a consistent way of handling the new projects. Presently we are working on having one position in the organization with the necessary knowledge and authority to be "system supervisor". This person will accept or decline the different suggested new applications, add on's and system changes. We also work on a standard formal project structure, to ensure that all small projects are quality controlled and according to the VIS philosophy.

We needs to find good criteria an application need to meet, in order to be included in VIS.
Presently applications with only one user and separate databases are included in VIS. VIS will become very complex, as more of these specialist tools are developed, if we allow all applications of this size to be included.

Today's VIS
VIS today help us in being an efficient customer oriented service provider. The customer center can answer most requests for data (Maps and installation documents) "on the fly" by using property-id or address as input to VIS. We also extract data for sale by using VIS. This is both pipeline information and geotechnical data. All in-house mapproduction, both standard and "one-of-a kind" thematic maps, are made by using VIS. All "in-house" employees" can make their own maps as needed, while we still produce analog maps for use in the field.

All document handling is now fully computerized. Incoming documents are scanned before being distributed. All correspondence out of the house is filed in the document system directly by the author. The document system can also handled reports, email etc. in addition to traditional correspondence.

The water and sewer module is used a lot in the "sewer" planning and also for daily maintenance work and rehabilitation. The water department is just starting to use the system for their planning and forecasting of flooding, snow measures etc, but do use it for their daily maintenance work.

VIS also provides the city's planning authority with accurate pipeline information, and geotechnical data. This is used both for planning purposes and building permits.

Figure 6 shows what VIS looks like today. As the figure shows we have not been able to reach the ideal world where data is stored and updated in only one database. We still have to use copies of the original databases in some of the applications. And we do not have online access to some of the external databases, like the city basemap database and property database. This causes some extra work, but fortunately the data updating and loading routines work quite well, so we do not have much problems with different versions of data in different databases.


Figure 6 VIS database and moduls - 2000


New applications 1996 -2000
We have in the period 1996 - 2000 added a new .Notification. system to the customer part of VIS. This has applications for Emergency handling, Construction work coordination and planned water shutoff. (The constuction coordination and shutoff appl. are replacements of original VIS programs.) The "Notification system" is used by the 24-hour emergency service phone personnel to register problems, and by the workforce to register follow up; i.e. what the problem is how it was solved, and when it was solved. It is also used to find which properties need notification in connection to planned interruptions in services, both due to construction work and water shutoffs. Both the application and the introduction of it have worked quite well, again I think, thanks to user participation in the development.

Another new VIS application is an add-on to the "office" system for complete digital document handling. (Also a replacement of a less functional earlier version in VIS) Here we have bought a shelf system. This application has caused some problems; it is not fully compatible with the basic philosophy of VIS; i.e. everything should be organized with geography as a common link. Although a very high percentage of our documents concern properties, the new document application is not built so that the property-id is registered and searchable. Our goal is to fix this as soon as possible.

VIS-data is presently used to do network analysis and planning work for water use. The modeling itself is done in separate specialized programs. VIS is also used for visualization and presentation of 3-D models. Today the pipeline network has to be "manually" built for network analysis, since our datamodel is a graphic and not a network model, at present.

The desktop mapping system is used for data presentations of various kinds. Water and pollution sampling results are visualized on the maps, giving good and accurate information to both the public and the city council.

Most of the functions specified in 1991 have been implemented in today's solution. The notification application was in a way our first .post. project application. And it showed us how much easier it is to get things up and running new when we have all the data, and also the infrastructure both for training and system integration.

If we look back to the project goals and ideas I think we can say that we have succeeded with VIS:
  • All "in- house" employees have access to the same updated information on their desk
  • VIS includes efficient applications to help OWSA reaching its strategic and business goals
  • OWSA today scores very high on both public service polls, and in giving service and helping the business community in Oslo.
  • Data entry and registration work is reduced by several man-years, and will probably be reduced even further.
The future - What are we at present working with?
Today we are mainly focused on efforts that can utilize the investments we have already done in the system. Below are the projects that we are presently running under the VIS umbrella.

  1. Customer services
    Improvement of the notification system for businesses is a high priority. When we know there will be construction work, or after we know of emergency shutoffs, we want to be able to notify the businesses that will be affected directly. Presently we only inform the property owners.


  2. Mobile units
    We have a pilot study running for field GIS solutions. The project will hopefully have 10 pilot installations by the summer 2001. We will only have data viewing in the field, no updating. We hope that the field units will save work and energy, by having all the correct and available information with them at all times.


  3. TV-inspection
    We have presently bought software for a pilot study on integrating TV-inspections in VIS. This will be done through the map module, and will include an integrated map based archive and viewing system for the videos from pipe inspections. This will make the videos more accessible and more easily usable for rehabilitation planning purposes.


  4. Internet customer services
    OWSA is proud of being a very customer oriented agency. We try to improve our customer services all the time. We are in the process of getting a web site for data inquiry up an running. Presently it will be restricted to our professional users. They will be able to order their maps, drawings and other documents on the web. And they can either receive the documents electronically, or by traditional methods.


  5. New versions of system for scanned documents
    We have problems both with our document handling system, and our system for scanned documents. The plan is to get new VIS modules for these applications during 2001. Hopefully we can keep the data formats as they are, without undertaking a major data conversion job.


  6. Integration localization in the economy system
    The new economy system in OWSA will enable the staff to localize expenses to a certain geographic position. I.e. adding GIS dimensions even to the economy system. This will help us in keeping track of which part of town has the more costly rehabilitation, and get the project cost of larger rehabilitation project divided according to localization. The economy system is ready for testing January 2001.


  7. Integration with laboratory database
    We plan to make a more permanent easy use connection between the lab.database and the desktop mapping system. This is to enable easier presentation of lab results from field samples.


  8. Organizational issues
    The VIS system today is working fine, but it is difficult to keep the system with all its applications updated at all times. We also have a problem with users creating their own subsets of data etc. Therefor the central VIS persons need to get focus back on the core system and data, and not necessarily participate in all the VIS subprojects. (As they do now) We also see that the company policy of distributing the GIS knowledge to each division only has worked to a certain extent. It is very difficult to keep developing new methods and new GIS products when everybody has to start from scratch. We are therefor in the process of upgrading the Mapping section to .Geospatial Information Section., with a separate function that will coordinate GIS use and effort throughout the whole organization.

  9. General system maintenance
    As the system sketch (figure 6) shows, the system utilizes several databases. We presently operate with copies of the public property database and the city basemaps. We hope to establish proper online connections to these databases in the future. We even copy our one utility map layer to a separate database used by the desktop mapping system. In the future we hope to be able to have one geographic database. Hopefully we will not need to copy any data, but can access all both "internal" and "external" data online from the source databases. Another challenge even further ahead, is changing our datamodel from a graphic model to a true network model.
We also hope the web technology will help with increasing the use of the system even more.
© GISdevelopment.net. All rights reserved.