next up previous
Next: Multiclass Equilibrium Auto Assignment Up: EMME/2 NEWS 13 October 1992 Previous: Time Stamps

Extra Attributes

One of the major new developments in Release 6.0 is certainly the introduction of the extra attributes.

For a long time some users have been asking for more than three user data items. With this feature, a user can now define a virtually unlimited number of network attributes for each scenario in which he can store any kind of numeric data that is relevant to his application.

Extra attributes are available at the level of nodes, links, transit lines and transit segments. Each attribute is given a name ('@' followed by up to 5 alphanumeric characters), a 40 character description, and a default value.

All extra attributes of one scenario share a common space pool, the so called extra attribute table. Its size is specified in words per scenario and is defined at data bank creation time or, for existing data banks, when the extra attribute table is created. During the life cycle of a scenario, this space can be reallocated as needed, by deleting extra attributes that are no longer used and creating new ones. Since the space needed for an extra attribute depends on its type (node, link, line or segment), the number of extra attributes that fit into a given table size is limited only by the total space used and by the grand maximum of 100 extra attributes per scenario (which should be enough even for the most demanding applications). A new module, 2.42 ``extra attribute manipulations'', is available in Release 6.0 for adding and deleting extra attributes, or changing their characteristics.

Each scenario has its own independent extra attribute table, which is empty when a scenario is initially created. When an existing scenario is copied, the extra attribute table is copied as well.

The extra attributes form a different class of data items, which is not part of the basic network definition. This means that they are not entered, edited and output along with the basic network data, nor are they subjected to the flag which protects a scenario from modification. In this respect the extra attributes are similar to assignment results. However, contrary to the assignment results, they are not volatile but their values are kept, even when the network is modified. Of course, if an element is deleted from the network, the extra attribute values associated with this element will be lost (but not all the others). On the other hand, if a new network element is created, the associated values of the extra attributes will be initialized to their respective default values.

Each extra attribute has a time stamp associated with it, which is set initially when the attribute is created, and then is updated whenever the attribute values are modified. Since this time stamp is displayed, along with the attribute description, whenever an extra attribute is accessed, it can give valuable information to the user as to the ``history'' of this attribute. Whenever the values of an extra attribute are modified, the user is also given the possibility to change the attribute description, so that the description will always stay in sync with the actual contents of an attribute.

Once defined in module 2.42, the extra attributes can be accessed and used in many different ways, as is shown in the following list:

As you can see from the above list, extra attributes can serve well in many cases where user data items had to be used up to now. Having the possibility to use extra attributes will allow all user data items to be used for storing information which is really part of the basic network definition, such as values that are used as parameters in volume-delay or transit time functions.

When creating a new data bank, the user is prompted for the size of the extra attribute table. Extra attributes can also be added to existing data banks. This can be done when calling the new module 2.42 for the first time with a data bank that does not yet have an extra attribute table. Since adding this table will increase the size of the data bank, make sure that enough space is left on your disk.

One of the primary purposes of the extra attributes is to contain derived data or results of application specific analyses, which are usually obtained with the network calculator. Color indices, level of service or emission of various pollutants are just a few of the possible uses. Also, extra attributes are an ideal place to store any kind of observed data in the base year scenario. These might include traffic or passenger counts (they can even be stored on an hourly basis, if needed), times, accident statistics, inventory numbers, or any other available data related to the network. This kind of data would normally be available from an outside source and, after having created the necessary attributes with the new module 2.42, be read into the extra attribute vectors using the batch input option of the network calculator, module 2.41.


next up previous
Next: Multiclass Equilibrium Auto Assignment Up: EMME/2 NEWS 13 October 1992 Previous: Time Stamps


Heinz Spiess, EMME/2 Support Center, Thu Jun 6 14:36:07 MET DST 1996