CASA: Commercial Strategy for a Geospatial Tool
This project developed a commercial adoption perspective for an emerging geospatial tool. My contribution integrated market scanning, academic network mapping and technical modelling. A central element was a network based simulation that projects potential user growth across universities. Monthly transmission probabilities along edges were derived from the observed average collaboration frequency between institutions (2020 - 2023), allowing adoption pathways to reflect empirical structural relationships rather than arbitrary assumptions.
The Data: Bibliometric Analysis
The starting dataset consists of citation derived institutional linkages forming a collaboration graph between UK universities. A sample of the raw publication records is shown below.
| Authors | Title | Year | Cited by | Affiliations |
|---|---|---|---|---|
| Loading data... | ||||
University Populations (HESA)
HESA staff and student counts were filtered to humanities related disciplines (English, History, Geography, Linguistics, Arts and related fields). These discipline specific counts weight each node, constraining the simulation to the most relevant academic communities for early adoption.
| University | Staff (Total) | Students (Total) |
|---|---|---|
| Loading data... | ||
University Locations
Geographic coordinates for each university are used to place nodes on the network visualisation and enable spatial reasoning about potential diffusion pathways.
| Name | Latitude | Longitude |
|---|---|---|
| Loading data... | ||
The Network: Mapping Institutional Connectivity
A Python workflow was developed to extract the collaboration network from authorship data, clean and associate with universtiy locations. Network edge weightings were also calculated to reflect collaboration frequency and provides probabilistic basis for the adoption simulation.
The Simulation: Projecting User Growth
The simulation estimates monthly staff and student adoption over 5 years with a 1 month timestep. Edge weights define probability of universities collaborating and hence potential for tool propogation. Within a seeded university staff adoption grows using the selected growth rate. Once the staff threshold is reached student adoption proceeds deterministically: 10% of remaining potential student users per timestep until saturation (for each university).
Instructions: Adjust the staff growth rate and student threshold then select Run Simulation. Use Pause to freeze the current state.
Staff Growth Rate: Controls per timestep probability of each eligible staff member adopting once the institution is seeded (current value 0.05).
Logic Summary: Monthly seeding: adjacent seeded institutions trigger adoption attempts along edges using collaboration derived probabilities. Staff adoption accumulates stochastically until the threshold (60%) after which student adoption proceeds deterministically at 10% of remaining potential per timestep.
Simulation results will appear here after running.