I wanted a repeatable way to test redevelopment ideas on a suburban commercial parcel. The area included an underused retail site, a divided arterial, residential streets, and a shallow drainage channel. I used PlaceMaker, Halfmaps, Shapezo, InfraWorks, and OpenRoads for separate steps, with a saved study boundary to make each pass easier to review.
1. Define the extent and the question
I started by writing down the decision I wanted to explore: could a portion of the parcel support housing while keeping clear access from the arterial and the neighborhood street? I selected an extent large enough to include both access points, nearby homes, and the drainage channel.
I used Halfmaps to review the mapped area and identify the layers I needed for orientation. I then used PlaceMaker to bring mapped context into SketchUp for early design work. I recorded the boundary and the source dates with the project notes. I did not treat mapped building edges, terrain, or imagery as authoritative measurements.
2. Generate a spatial draft
In Shapezo, I located the study area on the map and drew a rectangle around the parcel and its edges. Shapezo uses the selected map area as the scope for AI model generation. I saved the generated version alongside an image of the selection so another reviewer could understand what area the model represented.
The model helped me see the parcel, street pattern, open space, and nearby building forms as a three-dimensional scene. I used it to ask whether the study boundary was clear and whether a possible block arrangement deserved more attention. I did not use it to infer verified parcel dimensions or road grades.
3. Review the context
I brought the concept into InfraWorks and checked it from consistent viewpoints: an oblique aerial, the arterial approach, and the closest residential edge. I looked for visual conflicts such as a tall wall facing houses or a mass that blocked the view of an important intersection. Fixed viewpoints made the comparison easier when I changed the concept.
I kept the camera at a consistent height for each street view and compared the same intersections after each adjustment. That small discipline made the visual review more useful. Without it, I could have mistaken a camera shift for a design improvement. I also kept the aerial view because it showed whether the proposed connections actually lined up across the parcel.
4. Check the road design
I moved the access question into OpenRoads. I reviewed the proposed connection, turning movements, roadway alignment, and the relation between the entrance and nearby intersections. The specific design requirements depend on the project and the data available, so I kept this as a design review step rather than claiming the concept was traffic-approved.
Where road geometry depended on surveyed grades or traffic analysis, I flagged those inputs for the appropriate technical review. A clean model cannot supply missing source data.
For repeatability, I kept the access concepts in separate versions. One tested a single signalized entrance; another tested a connected local street. I used the same map extent and review cameras for each so the comparison stayed focused on the layout. If I changed the boundary and road design at the same time, it would be harder to tell which change affected the view.
For repeatability, I kept the access concepts in separate versions. One tested a single signalized entrance; another tested a connected local street. I used the same map extent and review cameras for each so the comparison stayed focused on the layout. If I had changed the boundary and the road design at the same time, it would have been much harder to tell which change affected the view.
5. Record uncertainty at every handoff
My handoff notes included the study extent, map sources, model date, and a short list of features that needed verification. I labeled the Shapezo output as an AI-generated early model. I kept it separate from the more detailed road work so later reviewers could tell which geometry was conceptual.
Why I keep the workflow modular
The sequence made iteration manageable because each step answered a different question. PlaceMaker helped me use map context in SketchUp. Halfmaps helped me read the selected area. Shapezo converted a map selection into a quick AI-generated model. InfraWorks showed how the concept fit its surroundings, and OpenRoads let me focus on roadway design.
When the study boundary changed, I could repeat the early steps without losing the road review notes. That is what makes the workflow useful to me: I can move quickly while keeping assumptions visible and design checks tied to their inputs.
