This paper explores to what extent inequalities in travel times, measured via the Gini index, depend on the spatial scale at which (average) travel times are measured. By using the new concept of Dedicated Accessibility Points, for the Netherlands we calculated average travel times at four spatial levels, ranging from virtually individual addresses to the level of municipalities. Travel times by car and bicycle to three medical points of interest are calculated: pharmacies, family doctors, and hospitals, as well as travel times by car from three other points of interest: ambulance stations, fire stations and police stations. At the level of individual addresses the errors made due to spatial aggregation is absent, but at higher spatial scales it plays a role. The results show that the Gini index is heavily influenced by the spatial scale at which the indices are calculated, with smaller indices at higher spatial scales. We discuss the implications of these differences for research and policy.@en

This paper examines the use of big data and data analytics in international transport networks from the perspective of historical big data, focusing on shipping logs from the British, Dutch, Spanish and French fleets in between 1662 and 1855. Based on a large-scale database containing mainly meteorological data collected in the CLIWOC project (2003), we computed travel distances and analyzed historical global maritime networks. This paper focuses on route choice, and consequently the time, distance, speed and reliability of the ships, covering different time periods, seasonal patterns and trade flows. The results reveal a clear picture of the main routes per nationality that is also indicative of the linguistical, cultural and economic colonial heritage that remains in the ‘host’ countries up to this day. The average daily distances covered vary over the countries involved, over the seasons and over different time periods. Also the trip characteristics vary notably over the different countries. Zooming in on the main trade flows, the corridor from the Netherlands to Indonesia stands out, but also considerable differences in average speed and stopover times were found along this route. Related to the complexity of using big data in studying international transport networks, our conclusion is that the degree of permutations and interactions with the dataset is not necessarily less for analyzing historical shipping records. It seems that big data of the past still can inspire future explorations of our historical transport networks on the world's oceans.@en