Statistical Modelling the Inventory Allocation of Bay Area Bike Share

Abstract

The project investigates the allocation of public bikes under the Bay Area Bike Share (BABS). BABS is the region’s bike sharing system with 700 bikes and 70 stations across the Bay Area surrounding San Francisco. It provides a convenient, environmental friendly, and affordable option for residents and tourists to transport within the region. Bay Area bikes can be rented from and returned to any station in the system, creating an efficient network with many possible combinations of start and end points. The main purpose for this project is to provide a more efficient solution for the Motivate, the operator of the system, to better allocate its inventories since it has many potential risks. We will build a generalized linear model with Poisson distribution by six variables: temperature, humidity, cloud cover, events, wind speed and visibility to predict the usage of bike per day. We believe that the company will achieve better revenues and avoid inefficiency by better allocating its resources since the control of cost is a critical aspect for business. At the same time, our solution will also be beneficial to residents and visitors who consider biking an option of transportation by matching bike supply with demand.