Beyond the hopefully surmountable #4, a bigger problem is perhaps likely to be the estimation of employment densities, for which building heights are also useless (also shown in script. What is probably going to be needed is a simple global model of demographic proportions, workplace engagement, and the likes. The OSM commercial and industrial buildings could then be used to locate and estimate densities of employment. This is nevertheless likely to be considerably trickier and more error-prone than mere estimation of residential densities.
Robinlovelace
commented
7 years ago
Beyond the hopefully surmountable #4, a bigger problem is perhaps likely to be the estimation of employment densities, for which building heights are also useless (also shown in script. What is probably going to be needed is a simple global model of demographic proportions, workplace engagement, and the likes. The OSM
commercialandindustrialbuildings could then be used to locate and estimate densities of employment. This is nevertheless likely to be considerably trickier and more error-prone than mere estimation of residential densities.