New in Emme 4.2

Our biggest 'minor' release ever, with new model workflow and dashboards, new transit modeling capabilities, an integrated spatial and relational database, more multithreaded procedures, improved performance, and core upgrades like longer names and more traffic classes, as well as included tools for interactive, scientific computing that go beyond traditional transportation forecasting applications. Some highlights below:

Emme Notebook

Introducing literate, visual and reproducible transportation forecasting and transport data science application development.

The Emme Notebook is an interactive computing environment for travel demand modelling and transportation forecasting applications. Combine Emme Modeller tools and Logbook entries, Emme Desktop worksheets and tables, Python code execution, rich text, mathematics, plots and rich media, as shown above.

Drag-and-drop model tools or maps and charts to create visual model dashboards that can be re-run at any time to keep results up-to-date.

Or author sophisticated transport modelling applications in Python with the Emme APIs, and bring modern data science to your transportation applications with the included SciPy stack, a great foundation for solving a variety of scientific and technical challenges.

Then share entire models with other Emme users in a single notebook document or with anyone else as a simple web page.

emme notebook

transit journey levels

Transit Journey Levels

Journey levels provide a new modelling capability to vary the generalized costs experienced by travelers along their transit journey. Explicitly model deeply integrated transit fare schemes, reduced waiting time at initial boardings, or force must-use transit mode(s) for combined-mode transit skims.

Journey levels are drawn from our recent research with strategy-based transit assignments, and preserve the well-known performance scalability and results integrity of the approach. Most important, unlike other approaches to modeling transfers through a transit network, journey levels are not limited to the transfers occurring at a local transit stop or station - multimodal transfers are considered network-wide.

Data Tables

Work with relational and spatial data directly inside Emme.

Save Emme network and demand data as data tables, and use standard relational and spatial operations like join, summarize, spatial join or buffer to perform analysis or compute new tables. Re-project between spatial coordinate systems. You can even write your own SQL queries and save results. Import tabular or spatial data from popular sources and map it. Or relate data table columns to use in model workflows as Emme attributes.

data tables2

multithreaded matrix

Multithreaded Matrix Calculations

The Matrix Calculator tool is now multithreaded, allowing speedups from 1.5x to 30x depending on number of threads, available RAM, matrix size and expression details. Choose number of processors and allocate RAM to play nicely with concurrent runs or other processing.

Even faster path-based traffic assignment

Delay functions in the path-based traffic assignment can now be precompiled yielding further performance improvements. Our benchmarking has shown 15-60% improvements, depending on application details like delay function complexity, size and demand characteristics. User-defined delay functions can be automatically converted and compiled.

Path-based traffic assignments

Longer names, and more traffic classes

Finally! Matrix and extra attribute names are longer, up to 40 and 20 characters respectively, so you can properly disambiguate all that data! (Recall you now can use up to ~40,000 matrices per database and 1,000 attributes per scenario).

Also, the Standard, SOLA and Path-based traffic assignments now support specification of up to 240 traffic classes for better value of time representation in toll models and other applications.