Transport infrastructure projects are typically appraised on the basis of travel costs and demand predictions from a traffic model—using the so-called rule-of-a-half to approximately measure the change in consumer surplus as an area under the demand curve. Alternatively, when the traffic model is a standard discrete choice model, the change in consumer surplus is equal to the change in the expected maximum utility of the discrete choice model, which in turn has an interpretation as a change in accessibility. When the traffic model has the nested logit form, the change in accessibility can be calculated in terms of so-called logsums.