Beta:
ModelBuilder is currently in beta. Share your experience and seek support through the forums available in the ModelBuilder Early Adopter Community.
Once the model has been built and contains at least one tool in the ready-to-run state, you can run the model. The sequence in which the tools run is based on the order the tools were added to the diagram and the order of connections between tools.
Note:
You must be connected to an analysis session and the model must be active to run it.
Run a model
To run a model, click Run 
on the model toolbar.
Note:
If the Run button is not available, it may mean that none of the tools in the model are in a ready-to-run state. Open the tool dialog box for a tool and verify that all required parameters are set.
All tools in a ready-to-run state run in succession. The model stops running when all ready-to-run tools have completed successfully or there is an error while running one of the tools.
The status of the tool is indicated with the following icons on the tools and output datasets:
- Success
—The tool ran successfully. - Warning
—The tool ran successfully but returned one or more warning messages. - Failed
—The tool failed.
Use the analysis history for the model run to review warning and error messages.
Rerun a model
There are several reasons why you may need to rerun a model, including iterating on an analysis workflow using different parameters, rerunning an analysis for a different study area, or fixing tools that failed in a previous run.
The following are the options for rerunning a model:
Run only ready-to-run tools
Clicking Run on the model toolbar runs all tools in the ready-to-run state, including tools that have been updated since the last model run, tools that failed in a previous model run, and tools that are downstream from a tool that was edited or failed.
Tools in a not-ready-to-run or has-been-run state will not run.
This option is best when you only want to run the updated tools in a model. For example, a model with three tools ran with a failure on the last tool. The first two tools ran successfully and produced the desired outputs. Updating the parameters for the third tool and clicking Run will rerun the third tool without rerunning the first two tools.
Run the entire model
Running the entire model resets all the tools in the has-been-run state to a ready-to-run state and runs the model from the beginning.
This option is best when you want to rerun tools that have run successfully, but the tool parameters have not changed. For example, you create and run a model so all tools and outputs are in a has-been-run state. While reviewing the results, you realize that a feature is missing from the input dataset. You can edit the dataset to add the missing feature, and use the Run entire model option to reset all the tools and rerun the model using the updated data.
Note:
Intermediate outputs from previous model runs are deleted when you run the entire model. Consequently, running the entire model may take longer than running only the tools in a ready-to-run state.
To run the entire model, complete the following steps:
- Click the Run down arrow
to expand the menu. - Click Run entire model.
Analysis history for models
When a model is run, an entry is added in the web map's analysis history. Individual tool runs in the model are not added to the analysis history.
Once a model completes, you can click the history entry to open the history details, and access the following tabs:
- Results—All results generated during that particular model run are shown. If a model is rerun without changing any parameters, new results may not be generated if the previous results are still valid.
- Messages—Error and warning messages for the model run as a whole, as well as for each tool run within the model, are shown.
Cancel a model
To cancel a running model, click Cancel 
on the model toolbar, or click Cancel analysis 
in the History pane. Any results that were generated before the model cancellation will still be available.