GIS Project Workflow?

Question

If you had to teach someone the key stages to a typical GIS Analysis project, what would be the best generic answer?

I would like to see diagrams to illustrate this point. Do not think it requires a long-winded answer.

I have been looking at ESRI's answer. This is the best I have found so far, but was wondering if we (community wiki) could do better. Im thinking, perhaps a nice workflow diagram would be good?

Feel free to edit Q, if it could be phrased better.

Answer 1

Instead of searching for a "typical" GIS workflow, we're better off looking for ways that GIS can fit into enterprise workflows. GIS Analysts often have a reputation for building their own information silos. We need to focus on ways we can package our analysis so that it plugs into an enterprise workflow.

Take Windows Workflow Foundation (WF), for example. As far as I know, Geocortex Essentials is the only GIS package that leverages WF. Essentials exposes workflows as REST endpoints. Like Silverlight/WPF, WF uses XAML which makes it easier to build designers. Part of the Geocortex Workflow is the visual designer, that allows you to simplify the process of connecting events and directing the actual workflow. You can see a screenshot on this Geocortex Blog post from August 2010. The designer embeds the WF designer, and so it looks similar to this:

David Chappell has a good discussion of WF here.

Answer 2

GIS Workflow in ArcGIS in simple terms

Workflow Manager schema now replaces the JTX (Job Tracking) 3.0.1 Database

Answer 3

Not sure whether it is what you are looking for (not exactly diagrams) but one of the best examples of GIS spatial analysis workflow I have seen is presented in this video:

http://www.youtube.com/watch?v=mMaMmaTfsQE

a bit fast, but shows how several transforms progressively reveal information from data. Other videos of the author may be relevant too.

Answer 4

I like the ESRI answer but would add some points.

1. You cannot assume that the data is in a form that can answer your questions in a simple way. Often the difficulty is converting a large dataset in to a form in which a geoprocessing tool or spatial SQL can work with. This is where GDAL, OGR, FME, Python and other such tools are so essential.
2. As a final stage it is essential to check your results using a different process. Outputs always need to be tested. This can be simply measuring with the ruler.