Abstract
M.Tech. (Operations Management)
Living in the current century, conducting interviews and carrying out field surveys is no longer enough. In an era, where everything has become smart, from smartphones to smart cities, a demand for smart analysis techniques has risen. Currently, knowledge gaps still exist in travel demand management (Giaimo et al,. 2010), hence a bridge is still needed to link what is available (big data) and what could be done (planning). “Advantages of applying smart technology to collect analyse data leads to flexible decision making as opposed to traditional cumbersome techniques” (Mokoena & Musakwa, 2016 p78-79). As no one model can be used as a one glove fit all situations, a need to continuously develop and renew planning models is essential.
This research reports on the spatial distribution of the Gautrain commuters, based on spatial predictions of the location of posts made on web 2.0 between the periods of January 2015 to June 2016. The findings from the content analysis highlight which train stations attract the most commuters and also possible locations for the expansion for Gautrain. In the study, the focal statistics presented the most visually accurate means of identifying clusters within a set radius. A hot spot belt was identified in areas near existing stations such as Park Station; Sandton; and OR Tambo, this which concurs with the commuter tag data from the Gautrain. Also, new hot spots were identified in areas which are currently not serviced by the Gautrain such as Soweto and Randburg in Johannesburg; Germiston and Alberton in East Rand; Montana Park in Pretoria. Similarly through the results from kriging, hot and cold spots are easily identifiable. Locations with hot spots should be further invested into by improving connectivity levels, as these are clearly points of interests for the commuters.
Future studies could run the model incorporating other control factors to determine variations using a time-series analysis, to identify any variations in hot and cold spots over time, thus areas which would present a constant hot spot would clearly be worth investing into. In conclusion the research presents a set of prediction tools to generate maps from web 2.0 posts to visualise and demarcate the various nodes of the Gautrain. These predications have proven efficient for a big data however, a drawback also arises, as the standard error becomes greater if small amounts of records are used.