| Home | GIS |
This paper deals with the management of a corporate land information system run by Milton Keynes Development Corporation. It explains why the system was judged to have achieved its original corporate aims. These aims are detailed together with the workload required to achieve them.
The paper also mentions the benefits experienced by the organisation and explains why the system is judged as a successful financial investment.
The paper then moves on to discuss some of the team management issues involved yet often overlooked within the GIS community.
Ken Robinson, a maths graduate, was a long serving officer of Milton Keynes Development Corporation. He managed the Land Information Systems project on behalf of the Corporation from the acceptance of the initial feasibility study through to the wind up of Milton Keynes Development Corporation on 31 March 1992. He is now employed by the Commission for the New Towns to manage the Land Information System inherited from the Development Corporation.
Many papers which have been circulated on the implementation of GIS within an organisation dwell on the mechanics of the system selection process, others concentrate on perceived academic advantages of one system over another; this paper considers the management of the total project in the light of the conditions which applied at Milton Keynes Development Corporation (MKDC).
Within IT many systems are only ever reviewed at the post implementation audit then management focus moves on; this is far too early to judge the success or otherwise of a GIS. For a GIS to realise the advantages which are identified in the feasibility study, a great deal of data requires to be captured. The cost of this data capture will far out weigh the installation costs for the system. The management of the data capture and, more importantly, of the team performing the data capture will have a great influence on the total time necessary to achieve a workable data set and thus on the success or otherwise of the system.
Very few organisations install a GIS without having some specific or implicit goal in mind; very often the fulfilment of the goal is some years away because of the requirement to capture data. The realisation of this goal should therefore be the yardstick against which the success of the system is measured; in commercial terms it is irrelevant whether the system is object oriented or employs some other currently favoured methodology provided it achieves the corporate aims. This paper looks at the reasons that the MKDC LIS was a success together with the author's view as to why this system has achieved its aims.
Milton Keynes Development Corporation (MKDC) was formed under an act of parliament in 1967 to develop a new city for 250,000 people in Buckinghamshire, England. After twenty five years of successful development, on the 31 March 1992, MKDC was wound up by central government and the responsibility for the continuing development of Milton Keynes was handed to the Commission for the New Towns, a government agency.
A development corporation will, during its life, amass a large housing stock together with an equally large commercial portfolio. During the period immediately preceding wind up, many of these assets are passed to various public and private bodies. In order that this process can be carried out quickly and accurately, access to up to date information throughout the disposal process, is crucial.
During 1988, two of the directors of MKDC, became increasingly aware of the problems which could be associated with the disposal of the corporation's assets. They instituted a study which eventually recommended the implementation of an LIS at MKDC, this system would ensure access to the most up to date data. As the initiative came from the top of the organisation the new system later benefited from powerful 'patrons'; in retrospect a feature which should not be dismissed.
The data collected, in the event approximately 120,000 polygons which were defined by the legal boundaries of 40,000 transactions, would represent the total land area which had passed through the corporation's hands and would form a 'definitive disposal plan' for the corporation's assets. The system would also enable the production, within the timetable set by the executive, of the myriad disposal plans required for the legal documents.
From the above it can be seen that the collection of a large amount of data had to take place within a comparatively short timescale. This timescale therefore dictated the programme for the implementation and data capture and also had a large influence on the style of project management.
The study also reported that the implementation, including system acceptance, could be completed by August 1989, effectively seven months later, leaving about two and a half years to capture the data and maximise the use of the system.
Very few authorities entering the GIS field possess the necessary skills needed to fully specify the hardware and software required for the implementation. Where an authority lacks these skills it is false economy to proceed without buying those skills from outside. In consequence MKDC formed a fruitful relationship with the GIS manager of an I.T. consultancy. With this assistance MKDC were able to draft a functionality specification, as a result the tender did not specify the system configuration. Prior to drafting the tender, there had been an executive decision that MKDC would only implement a system which was tried and tested and could quote at least four U.K. reference sites. This gave a strong lead to the implementation team in that we were not looking for a 'state of the art' system but a reliable proven system which was already in existence.
Once tenders had been returned, selected tenderers were supplied with a set of typical data to allow them to prepare a prototype system for assessment. Their ability to complete the prototype within a very short period was seen as an indication of their ability to meet MKDC's fast track programme.
In order not to become overawed with the 'bells and whistles' that the vendors stress, a weighting was attributed to various criteria prior to the assessment of the systems. Because of the need to collect the data concerning the 40,000 legal transactions, we placed a very high emphasis on ease of data collection and a comparatively low weight on spatial analysis. As a result of the weighting the selection team unanimously chose one system to recommend for implementation.
The negotiation which took place with Intergraph (UK) prior to the contract being signed, was perhaps the most fruitful part of the pre-installation phase. As all the tenderers had been treated in a formal but friendly manner during the tender phase, this was the first opportunity to build a good working relationship with them. The author felt it was important, during these meetings, for their staff to be able to gauge the spirit in which the contract would be administered and to realise that a pragmatic approach to the implementation would be adopted.
After the contract was signed both the hardware and software installation was completed within the contract period.
The decision was taken to centralise all the graphic and alpha-numeric terminals within one dedicated LIS section. This arrangement led to tighter control of the system and enabled an interchange of ideas and methods during the crucial learning period. A further benefit was realised in that the system was staffed by a recognisable team of people.
During the rest of 1989 data collection proceeded and it was soon apparent that work flows had to increase to cope with the daunting data collection task. After examining various options it was decided to operate the equipment ten hours a day seven days a week using two teams working a four day on, four day off rolling week. By employing this strategy, workstation utilisation would be doubled with, it was hoped, comparable gains in the data capture programme.
The beginning of 1990 not only saw the introduction of the extended working but also a marginal increase in the number of workstations. As a result a number of new operators were required, the team very quickly needing to double in size. The additional staff were only required for 27 months, in consequence an agency was employed to provide these staff. It is also important to realise that additional staff were required to support the operators carrying out the data entry. As a result of our rigorous checking procedures we required two staff to support each operator. Thus the project team numbered 23 people during 1990 and 1991.
In order that a GIS has any credibility within an organisation the data within the system must be reliable. In order to achieve this reliability a three stage checking process was carried out in MKDC, both on the text data and on the graphical data - hence the large team. Although this process was time consuming it formed the foundation on which the disposal boundaries were later to be determined.
Many papers have been written giving suggestions and advice on managing change within teams. Within the GIS fraternity at large, people spend a lot of time and money on the technical systems, however without effective management the new teams that are formed will fall apart and targets will suffer. I would suggest that project leaders don't have to slavishly follow any of the guidelines available for the successful management of projects; but they should be aware of the effect on morale and team spirit if they chose to adopt unrecognised management techniques.
Although MKDC placed great emphasis on good team leadership practices, it was inevitable that the necessities of the programme targets would require action which did not allow proper consideration to be given to all the team members. The project leader should be very aware when this is happening and try to ensure that harmony is restored as soon as possible. At times one person's contribution to the project becomes of overriding importance. It then may be necessary for the project manager to allow them to assume the day to day leadership of the team with consequential difficulties for other members of the team. However once this crucial period has passed then the balance should be restored without delay.
In order to achieve the corporate aims neither the team nor any of its members, should become side tracked into creating 'the best, most advanced system in the UK'. It is easy for this to become one of the unofficial aims of the department, perhaps only naturally when one considers the technical expertise and pride in the system which is evinced by the senior members of the team. They must focus on the task and only use that level of technical sophistication which is necessary to complete the task on time.
Within a large team there are a number of duties to be performed, however there are also a number of social functions that the team members will require from each other. The balanced team needs members who fulfil the functions of 'explorers, advisors, controllers and organisers'. Recruiting staff in the image of the project manager will ensure an unbalanced team. Within Information Technology generally it is easy to end up with a team of 'explorers' who are too involved with the sophistication of the system and not interested enough in the capture of data. Staff involved in the routine capture of data are often considered inferior to technicians, these staff require the same job satisfaction and career prospects as technical staff.
However the work groups are set up the staff will chose to form alliances and unofficial teams, often crossing group boundaries. These alliances can be used to advantage if you identify the natural and unofficial team leaders involved. For example one of the junior staff who worked on our project had a reputation as being difficult to manage. She was, in fact, the natural leader of one of the unofficial teams and acted as their spokesperson. If such a person is not handled with sensitivity and recognition of their unofficial role then the whole team will become unsettled with consequent problems for the project.
When the department is working flat out, it is important to take time out from the immediate task and stop to review the situation. It is helpful to look back over the recent decisions that have been taken in order to confirm that the project is moving forward along the optimum path. This can be carried out in a formal manner using critical path analysis and other such tools, or it can be an informal process whereby the senior members of the team gather once a week or as necessary and review and challenge the recent decisions taken.
As a part of this on going review process it is important to involve the team members in the project. Obviously all the team cannot be part of the decision making process, but it is important that they are informed of all decisions taken as soon as possible. A project leader will get more support from the team when difficulties arise if the team are accustomed to regularly receiving an honest appraisal of the project. When the project is not coming up to expectations but the team is working hard then the project leader should consider whether his targets were realistic or not. If the targets were not realistic then admitting this to the team and praising their efforts to meet the deadlines can spur the team to work miracles.
Obviously in order to staff such a project one needs to recruit a number of staff with varying skills. When MKDC tried to obtain sufficient trained staff they were not available. Hence inexperience staff were recruited and trained on site. This lack of trained staff also dictated our three stage validation process. In training staff we tried to impart as wide a knowledge base to as many staff as possible. This ensures that staff get proper career development and also management are not embarrassed if their only data base administrator leaves. Technical training also assists greatly with the on going process of team building.
As stated earlier one of the prime functions of the system was to ensure the speedy and accurate production of disposal plans. At the end of the life of the development corporation many of its assets were transferred to other bodies. The three major elements of the transfer comprised a total of 13,000 rental and equity shared ownership houses together with 168 kilometre of major roads and 2000 hectares of parkland and road corridor landscaping. Together with these transfers other assets were also being transferred such as play areas, commercial units, community halls and lakes. In total we were required to produce in excess of 9,000 disposal plans in five months.
During the most crucial period, the Autumn of 1991, it was imperative that we identified those skills possessed by individual team members. Only by doing so could we use their expertise to the best advantage of the project. Our team was split into smaller groups: One group was concerned with the checking of data and ensuring all information was up to date, one with the identification of the boundaries of the linear parks and transportation corridors. A third group was responsible for the capture of housing estate boundaries and individual disposals. As well as these three groups there were three people with specific responsibilities. One person was responsible for the integrity of the data and the liaison with the other involved departments. Another member was responsible for all the plot output of the department and managing the plot software and the plotting workload. Also a further member of the team assumed the role of trouble shooter, removing any hold ups which affected the smooth workflow through the section.
When planning this work it was realised that to produce the 9,000 plans which were required would necessitate the plotter working continuously, day and night, for nine weeks. This assumed no peaks and troughs in demand. Obviously this situation was going to place excessive, if not impossible, demands on both the team and the system. However it was realised that many of these plots would be multiple copies of an original. Probably only 2,000 plot requests would be submitted to produce these drawings. After investigating the possibilities, it was decided to sub-contract the duplication of the transfer drawings to a colour copying house who had a number of Canon A1 Bubble Jet colour copiers. These copiers were identified as giving good colour reproduction with minimal scale distortion for large sized plans. Where we required more than four copies of the transfer plan, we produced two copies in house and dispatched one copy to the Color Company for reproduction. By this means the in house production of the transfer plans was brought to a manageable level with the copy house preparing about half of the necessary plans.
The main constraint to the whole transfer process was the time scale imposed by the tenants right to vote for the landlord of their choice. The results of the tenants' ballot were not available to the department until the end of September; the transfer was scheduled to be complete by the end of the year. In consequence the plans were required to be completed by the beginning of December. In order to minimise the work carried out during this extremely busy period all the work to prepare the necessary drawing files was carried out during the ballot.
Post ballot, the information was passed to the department as a series of drawings showing which houses would be transferred to each landlord. As a further complication the development corporation were still selling, up to the end of November, some rental houses to sitting tenants under the right to buy legislation ; this meant that although a drawing could have been completed there was a chance that a house would not form part of the transfer. Obviously this process was open to last minute changes being introduced.
In line with all our work, and in order to minimise the errors on the transfer plans, they were subject to our usual three stage checking process. Initially the transfer drawing was digitised into the previously created file from a rough drawn original. A plot was then obtained and checked against the original. All discrepancies were marked on the plot and next these discrepancies were checked by a more senior person. The resulting edits were performed in the graphics file and a further plot then checked by the housing department prior to the production of the legal plans. This process was programmed so that the transfers were not delayed waiting for the necessary plans. The three phase error checking also ensured that the plans contained negligible errors.
One of the transfers involved estate related landscape assets and required multi-coloured shaded area plots to be produced as a result of spatial analysis. it was soon realised that an unconventional approach to the analysis would provide the most suitable production method. In the event a plot macro was written which, by manipulating the priorities of themes excluded the land which was not part of this transfer. As the base information was constantly changing up to the last minute, this methodology allowed the necessary analysis to be performed at the instant of forming the plot meta file.
The transfer of the assets of MKDC could not have been achieved during the remaining life of the corporation without both the Land Information System and the team of people who were involved in the data capture and preparation of the transfer plans. It is estimated that the transfer which took five months would not have been completed in under eighteen months using traditional methods.
Milton Keynes Development Corporation had a great deal of experience in managing multi-disciplinary projects and this experience was found to be invaluable during the system procurement and implementation. The knowledge required to implement any GIS is going to cross discipline boundaries and staff must be prepared to educate themselves within the new areas.
Remember that the total team extends outside the GIS department; it will include staff from other departments of the organisation. Also members of the various contractors staff will be part of the extended team.
The most important aspect of any successful GIS is the team who are employed on the project; without them the system bought at great expense just becomes a collection of expensive junk.
| Copyright © 2001 Ken Robinson |
