While most users are probably still in the process of discovering all the new features and enhancements offered in Release 5, the development work in preparation of the next release is already in full swing. In the following paragraphs, we'll try to summarize some of the developments that were carried out in the last few months and which are now about to enter the alpha test phase. Of course, many other development projects are also already in preparation; but, as always in this column, we stick to the rule of presenting here only work for which the basic implementation has been completed and which has already, or is about to enter the alpha test phase.
Literally for years, module 2.11, "Input base network using batch input", has lived the life of a neglected stepchild. Since the very beginning of EMME/2, it had been implemented in a "quick and dirty" way, which only allowed to read in a complete base network, with the additional requirement that the input file needed to be presorted according to node numbers. There was no possibility to add nodes or links to an existing network, even less to modify or delete network elements! No wonder, that nobody was happy with this module, and no users group meeting went by without a user complaining about it (and rightly so!). The reason for delaying this development for so long was not because we did not see its necessity, but simply because there were always other developments which had an even higher priority. Finally, at the last Users Conference in Pasadena, we made the firm commitment that 2.11 would be irrevocably the next item on the list of developments. As I am pleased to announce here, we stuck to our promise and we have now a completely rewritten version of module 2.11 which does away with all the former restrictions and omissions. This new version fully implements the add, modify and delete operations for links and nodes, and can handle these operations in any arbitrary order, even allowing for multiple nodes and links sections within the same input file. In addition, attribute keywords are now fully supported (e.g. for selective attribute modifications) and a special notation is provided to add or remove modes of the link mode attribute without explicit knowledge of the remaining link modes.
Another question which invariably came up at every users group meeting
was the lack of a possibility to increase the dimensions of an existing
EMME/2 data bank. Here too a solution is now in sight! It consists of
a new macro MODIFYDB, which is to become an integral part of the EMME/2
software distribution. This macro uses module 1.11 to perform changes
in the data bank dimensions, which were defined when the data bank was
created initially. The current implementation of the MODIFYDB macro
allows the following operations:
More operations are planned to be added to MODIFYDB at a later stage.
Note that even though the MODIFYDB macro is only scheduled to be
distributed with the next official release, it can be run already today
using the standard EMME/2 Release 5 software, on both DOS and UNIX systems.
VAX systems require Release 5.2 (see below) to run this macro.
One of the claims made with the introduction of the new network calculator in Release 5 was that this would allow the true implementation of the congested transit assignment based on a non-linear equilibrium model, as was described in Spiess and Florian (Transpn Res 23B, No 2, 1989). In the mean time, INRO has developed a macro which implements this assignment model by combining the standard (uncongested) transit assignment with the functionality of the new transit network calculator. This macro is now being tested with the aim of making it available to all EMME/2 users.
While only few users are in the habit of writing such elaborate EMME/2 macros as the one mentioned above, most users are relying on some macros to implement complex procedures or to automate repetitive tasks. Since its first introduction in Release 2, the EMME/2 macro command language has been steadily enhanced and can now be used to tackle almost any task within EMME/2 a user can think of. As more macros are being written and used, and the tasks they implement get increasingly complex, there is also a growing need for having access to even more elaborate macro constructions. For the expert macro writers among you, here are some highlights of the newest enhancements to the EMME/2 macro command language:
These enhancements pursue multiple goals. Firstly, they will, of course, increase the functionality of what macros can be used for, secondly, they simplify the work of macro writing by allowing a more direct implementation, and thirdly, they allow the macros to become much more "user friendly" by hiding any unnecessary dialog from the user.