ANL303: Shared e-scooter has emerged as an affordable transportation means for short-distance trips. It also helps alleviate traffic: Fundamentals of Data Mining Report,, Singapore

Shared e-scooter has emerged as an affordable transportation means for short-distance trips. It also helps alleviate traffic congestion by reducing the number of trips made by vehicles. However, e-scooter-sharing service providers face numerous operational challenges.

One of the challenges is to ensure that riders can always rent and park the e-scooters at stations and that the e-scooters can be sufficiently charged before the next trip. Ideally, at the start of a trip, stations should have sufficiently charged e-scooters for riders to rent; at the end of a trip, stations should not be full so that riders can park the e-scooters.

From an operational perspective, this requires service providers to send trucks to redistribute e-scooters from full stations to empty stations so as to have balanced stations. Some providers have been spending a lot of money to perform overnight charging and re-allocation of e-scooters among stations.

Assume that a dataset is collected from an e-scooter sharing service provider to perform data mining in an attempt to generate useful insights to solve the abovementioned rebalancing problem. The dataset contains details of each e-scooter trip made by individual riders. The variables in the dataset are described in Table 1.

• Give one (1) example of data quality issues that may potentially exist in the dataset described in Table 1 and propose a solution for it.
• (b) Based on the dataset described in Table 1, give one (1) example of descriptive data mining and discuss how it might generate useful insights to solve the rebalancing problem.
• Based on the dataset described in Table 1, give one (1) example of predictive data mining and discuss how it might generate useful insights to solve the rebalancing problem.
• Suggest two (2) additional variables that could be included in the analysis for solving the rebalancing problem. Explain the rationale for their inclusion and describe how you can collect them.
• When the rebalancing problem is not handled properly, one of the consequences is that some riders do not return the e-scooters to the designated stations if there is no empty parking slot at their desired end stations. Describe how this would bring negative impacts to society.