12.8 Prioritizing new infrastructure

This chapter’s final practical section demonstrates the policy-relevance of geocomputation for transport applications by identifying locations where new transport infrastructure may be needed.Clearly, the types of analysis presented here would need to be extended and complemented by other methods to be used in real-world applications, as discussed in Section 12.9.However, each stage could be useful on its own, and feed into wider analyses.To summarize, these were: identifying short but car-dependent commuting routes (generated from desire lines) in Section 12.5; creating desire lines representing trips to rail stations in Section 12.6; and analysis of transport systems at the route network using graph theory in Section 12.7.

The final code chunk of this chapter combines these strands of analysis.It adds the car-dependent routes in route_carshort with a newly created object, route_rail and creates a new column representing the amount of travel along the centroid-to-centroid desire lines they represent:

route_rail = desire_rail %>% 
  st_set_geometry("leg_orig") %>% 
  line2route(route_fun = route_osrm) %>% 
  st_set_crs(4326)

route_cycleway = rbind(route_rail, route_carshort)
route_cycleway$all = c(desire_rail$all, desire_carshort$all)

The results of the preceding code are visualized in Figure 12.6, which shows routes with high levels of car dependency and highlights opportunities for cycling rail stations (the subsequent code chunk creates a simple version of the figure — see code/12-cycleways.R to reproduce the figure exactly).The method has some limitations: in reality, people do not travel to zone centroids or always use the shortest route algorithm for a particular mode.However, the results demonstrate routes along which cycle paths could be prioritized from car dependency and public transport perspectives.

qtm(route_cycleway, lines.lwd = "all")

Figure 12.6: Potential routes along which to prioritise cycle infrastructure in Bristol, based on access key rail stations (red dots) and routes with many short car journeys (north of Bristol surrounding Stoke Bradley). Line thickness is proportional to number of trips.

The results may look more attractive in an interactive map, but what do they mean?The routes highlighted in Figure 12.6 suggest that transport systems are intimately linked to the wider economic and social context.The example of Stoke Bradley is a case in point:its location, lack of public transport services and active travel infrastructure help explain why it is so highly car-dependent.The wider point is that car dependency has a spatial distribution which has implications for sustainable transport policies (Hickman, Ashiru, and Banister 2011).