FIDICcharts uses a unique visualisation strategy for data repositories that builds upon modern web-based browser technolgies. Fig. 1 illustrates the strategy.

Fig. 1 - Visualisation strategy.

We deliberately use non-technical terminology (e.g., "tables") to describe the strategy. We introduce the approach to visualise and access the various hierarchical levels of information stored in a repository by using the terminology used for project sustainability rating systems that often speak of "dimensions", "themes" and "issues". A highest-level dimension is "governance", with "oversight" as a theme at the next level and several issues such as "anti-corruption programme" at the next level.

For a governance dimension, we have several Objects Tables that record Data Objects associated with each theme. A Theme Chart is created using the Theme Objects Table with a Theme Layout Table which specifies how the Data Objects for each theme should be laid out on the Theme Chart.

This discussion assumes that the Theme Chart will use swimlanes as the visualisation so the layout is defined in terms of swimlane concepts, namely the Actor (i.e., the Data Object's swimlane) and the Position in the swimlane. Other types of visualisation would use more complex positionning.

Higher and lower level charts are constructed in the same way, namely:

• all the themes are combined to give a Dimension Chart;
• a given theme is decomposed into Issue Popups (we use popups since in a browser implementation as the the amount of information is relatively small a popup instead of a chart is sufficient, and more efficient);
• two or more Issue Popups are combined to give Issue Cross-Popups.

Consider first an information repository with a simple structure. A good example is a sustainability rating system where there is only one actor, for example, the organisation managing the rating. There are dimensions, themes, issues and criteria for each issue. The criteria are scored. The information repository in this case has a simple structure:

• Dimension 1
  • Theme 1.1
    • Issue 1.1.1
      • Information (criteria, performance targets, evaluation methods)
    • Issue 1.1.2
      • Information (criteria, performance targets, evaluation methods)
  • Theme 1.2
    • Issue 1.2.1
      • Information (criteria, performance targets, evaluation methods)
    • Issue 1.2.2
      • Information (criteria, performance targets, evaluation methods)

In our approach, the dimension Chart is a row with series of Themes organised along one timeline; a Theme Chart is a row with a series of Issues organised along one timeline where the timeline corresponds to additional requirements corresponding to an increasing level of performance (see Fig. 2).

Fig. 2 - Sustainability rating: simple visualisation strategy for one-actor.

Typically for a rating system one would have three performance levels. A Theme Window scanning the Theme Chart would expand the theme's objects. Selecting a theme object would display the theme's issues in a popup allowing the data for each issue to be easily accessed (see Fig. 3).

Fig. 3 - Sustainability rating: simple visualisation for one-actor.

Frankly, viewing the dimension and theme timeline charts of Fig. 3 would not be particularly instructive for this very simply repository structure. Nor would it be particularly efficient in terms of the visual space required. Other visualisation approaches such as partition tables and tree tables are much more effective and efficient.

However, the situation changes dramatically if the repository aims to help several actors manage, access and interact with the repository to achieve different performance levels.

Take as an example the case of three actors, let us say the promoter of an infrastructure project, the design team and a stakeholder's representative, these being the principle actors interested in the sustainability of an infrastructure project such as a water treatment plant. Each actor will have a different role to play. The dimension chart could have a swimlane for each actor with action items for each theme organised in sequence according to the performance level.

For the Theme Chart (Fig. 4), the Data Objects are labelled 1.1a, etc with 1.1 corresponding to the theme and a, b, c .. corresponding to the action identifier. The Data Objects would be grouped under several performance levels (Performance Level 1, ...). A Theme Window scanning the theme chart would show those actions that the three actors need to address for a given performance level. Actions are usually linked (i.e., the output of an action by one party is the input for action by another party) and the Window Chart would display these relationships between the action items.

Fig. 4 - Sustainability rating: three-actor visualisation with performance levels.

The proposed theme chart would allow the pincipal actors on a project to see at a glance the issues and requirements that need to be met to achieve a given performance level for each of a ranking sysytem's sustainability themes. As such the visualisation approach would complete the visualisations being developed for the Project Sustainability Logbook (see website).

Sample letters, sometimes called standard letters, for FIDIC contracts are sent between the Employer, the Contractor and the Engineer.

A web application (see the Charts link above or click Charts) has been developed to help coordinate and track the various letters. The application uses swimlane charts to chart the sequencing of the various letters while a contract progresses.

In the case of a contact such as the FIDIC Construction Contract, the dimensions, themes, issues, and data equate to the following (the terms used by Hipp al - see Concepts - are given in square backets):

• Dimension = Principal Events (Area)
• Themes = Procedure [Refined Area]
• Issue = Sub-Procedure
• Data = Data [Data]

For the Objects and Layout Tables:

The Charts app provides a Contract Chart for the FIDIC Construction Contract. It comprises:

• Principal Events
  The lower Principal Events chart can be scanned. It gives:
  • main events actions by the three parties to the contract;
  • highlighted Principal Events that make up a selected Procedure.

• Procedure
  The upper Procedure Chart can be scanned. It gives:
  • Procedure events and actions by the three parties to the contract for the selected Procedure;
  • the way events and actions are linked, with time periods;
  • links to the chart for a selected sub-procedure;
  • popups of the data fo each procedure item.

The aim is shortly to adhere to the proposed visualisation scheme by having the the sub-procedure popup display both sub-procedures and data.

The visualisation approach described above is particulary well suited for data repositories that have several levels and a fairly high level of grannularity. For the examples given above, the number of data items is 5000 to 10000 (see Table 1) which can be easily handled by web-browser Javascript using JSON string arrays as opposed to a relational database that has a much larger, and probably unacceptable, performance and memory footprint if the in-browser application is to be responsive.

Table 1 - Elements per repository level.

Given that the contract demo website - see Charts - works well, one can envisage applying our visualisation approach to other forms of structured repository information: such as:

• rating systems
• sustainability logbooks
• standards
• manuals
• discussion forums

---

Please contact Peter Boswell at PeterBoswell.com

© Bricad.online, Bricad Associates Sàrl, January 2021