GISdevelopment.net ---> GITA 1997 ---> Best of the Rest

Schematics, as Output

David Fensterheim
Formula Software Services, Ltd., 35 Reiness Street, Israel 43374


Abstract
Public utilities such as telephone, electric, gas and water companies are using AM/FM (Automated Mapping / Facility Management) systems to manage their networks. AM/FM technology combines a graphic/ geographic interface with a relational database to manage a companies facilities and to track how cables and other facilities are connected. These systems shorten the time required to design new network and generate engineering work orders.

Traditionally an operator used an AM/FM system to produces maps and reports. In addition to maps, however, it is often necessary to produce details of how a network is connected. Schematic diagrams legibly display details of how items of plant are connected. In order to present these details clearly, the graphics is oriented logically rather than geographically. The schematic linework is typically drawn manually and connectivity information is entered during this process.

This paper presents an automatic schematic generator. A schematic generator not only saves the time required to draw the schematics but makes operations, such as connecting cables and inspecting connectivity, more efficient. Complex connection scenarios may be automated since all data entry is alphanumeric.

Traditional Schematics in Telecommunications (Outside Plant) Connectivity diagrams are particularly important in the design and management of communications networks. A telephone cable may contain any number of wires or fibers. Each wire pair or range of pairs/fibers maybe connected to a range in a connecting cable.

Imbedded Details
Various attempts are made to display the connectivity of a telephone network on a geographic network map. Details displaying the connectivity (pair counts, joints, etc.) of a specific location (e.g. hole, pole) are drawn on the map on or near the junction location. All connections occurring at the location are drawn on the detail.

In order to avoid considerable clutter, the details are drawn at a scale only legible when viewing the detail only. When displayed on a computer screen, the viewer must zoom in so that the diagram fills the screen. When plotting a map at a typical scale (1:500 or 1:1250) the details become illegible.

Off Map Details
To solve the clutter problem, these diagrams are often drawn off the map as separate details. In some systems, a continuous joint diagram of an entire phone network is drawn.

Tables
Where each cable is uniquely identified, a table maybe created which lists the wire/fiber assigmnents. At Bezeq’s northern district in Israel, a table for each joint is produced resembling the following chart. The number of columns correspond to the number of feed and fed cables. Tables for a number of joints maybe horizontally chained showing the continuity of a main cable or trunk.


This format was adopted into Bezeq’s AM/FM system as both a report form as well as a data entry form for defining connectivity. This method however does not provide a clear graphic diagram which would convey the information in a more legible format.

Problems

Manual Detail Entry
Whether a detail is drawn on the geographic map or off the map, they are usually entered into the AMIFM database by manually drawing graphics. The detail graphics may be linked to the geographic graphics thereby allowing the application to display the detail of a geographic object or visa versa.

In a typical placement scenario:
  • Strand data and cables are placed on the map
  • On a separate area of the graphics space, a detail is drawn of a manhole or pole and the associated joints.
  • Lines representing the cables (previously placed geographically) are drawn to the detail.
  • Lines are drawn connecting the cables during which time the user specifies the specifics of the connection (pair count etc.).
This process is both time consuming and restricting. It is time consuming since what would normally be simple data entry becomes a matter of drafting. It is restricting since connectivity must always be defined the same way.

The Schematic Generator

How The Generator Works


Figure 1. Generated Schematic

Typical AM/FM systems are capable of tracing a network once the Items of Plant have been connected. A trace maybe used to generate a report or modifi attributes of the features in the traces path. In the telecom case, pair counts maybe modified during such a trace.

To generate a schematic, the network is traced producing records of features contained in the network. The database is then queried for supplemental information such as address names etc. Once all the network information has been extracted from the database, a diagram is produced, graphically representing the traced network.

Advantages
Israel’s Bezeq with Formula Ltd. have developed an Outside Plant schematic generator. When designing or converting an outside plant network, an operator places only the graphics representing the location/geometry of the items of plant. No connectivity information is entered graphically or displayed on a map. This has resulted is extensive time savings in the preparation of details.

In addition, automatic connectivity fi.mctions may be developed so that complex connectivity scenarios may be automated. When converting existing network maps and records, connectivity information may be entered alphanumerically rather then entered while drawing graphics. In other words, no lines are drawn to describe how two cables are connected.

Examples:
  1. When connecting a feed cable to a number of fed cables pairs are automatically assigned based on the demand of each fed cable.
  2. Certain complex connectivity scenarios which periodically repeat, maybe automated, further reducing the time required to connect a series of cables.
Graph theory and Computer Algorithms

Tree Networks
In telecommunications, common copper networks fall under the category of a tree network or tree “graph”. Therefor by recursively sorting the traced data, all connections at a particular vertex are isolated and each vertex is assigned a location within the tree. The algorithm computes graphics of each vertex, the size of each vertex and the physical location on the diagram. A list of modifiable parameters are listed in an initialisation (.ini) file and the graph is drawn based on the parameters.

Recent enhancements to the generator allow the schematic to “fold” so that a schematic will grow down, rather than across the page. Traditional (hand drawn) joint diagrams cluster all joints which are located geographically at the same point (e.g. on the same pole/hole) at the same graph location. A tree graph will position each joint at a location based on the joints logical tree position. This limitation often requires utility technicians to get used to a new form of diagram. We found that the benefit of clear computer generated documents strongly outweighed the initial discomfort of learning the new format.

Fiber Networks and Redundant Rinm
In the case of optic networks, redundant ring technology is used and does not conform to the tree model. Therefor a specific solution was provided for the case of redundant rings. This solution was an adaptation of the tree network generator and not a full blown network graph. Since the classic telephony model is a tree (feed to fed) model, reverse connectivity was added to the model in order to describe the backup connections in a ring network.

Non Tree Graphs
An AM/FM schematic generator is a perfect application of graph theory. The next step in the development of a schematic generator is the implementation of a non-tree schematic.

This type of schematic will be able to produce:
  1. Traditional joint diagrams where a single vertex maybe defined as a geographic location such as a hole or pole.
  2. Schematics for non-tree utility networks (e.g. electrical).
  3. Better support of non-tree telecom networks.
For more information on graph theory, see http://www.math. fau.edu/locke/graphthe.htm

© GISdevelopment.net. All rights reserved.