Stations that ranked in the top 16% in four user demand metrics were identified earlier as top-performing stations. There were nearly 100 such top-performing stations.

Impossible to Use Market Penetration to Optimize Size of Top-Performers

The 16% fraction was chosen in order to get 100 top-performing stations. In order to explore the feasibility of determining an optimal number of top-performing stations, the fraction of stations with ridership in all years from 2018 to 2022 was varied from 10% to 100%.

The

  1. resulting number of top-performing stations
  2. 2022 market penetration (total ridership accounted for) by stations identified as top-performing stations

were then calculated.

Below is a chart showing the 2022 market penetration as the number of top-performers is increased from 10% to 100% of all stations used in bike share trips between 2018 and 2022

00_market_penetration_by_num_top_stations.png

The red line shows the number of top-performing stations (100) when the top 16% was considered as top performers. The smallest fraction of stations considered was 10%, and not zero. For this reason, the left end of the chart does not pass through (0, 0).

Below are the key findings from this chart

  1. In 2022, the 100 top-performing stations captured a 42% penetration. The next 42% (1X) of 2022 ridership was captured if an additional approximately 280 (or 2.8X) stations are included, which is a multiple of 2.8. In order to capture the full 58% (58/42 = 1.3X) of remaining ridership, the remaining approximately 530 (or 5.3X) stations are required, which involves a higher a multiple of 5.3/1.3 = 4.1X. The multiple is higher is because the market penetration chart is asymptotic and not linear. It captures greater market share more rapidly near 100% than near 0%.
  2. There is no evidence of a knee in the green line on the chart. If a knee were present, then this location could be used to identify the optimal number of top-performing stations: below the knee (starting from 0) market penetration would increase faster as more stations are included than it does above the knee. This vehavior is not present in the chart since the green line displays the asymptotic behavior mentioned above. As a result, it is not possible to use this chart to determine the optimal number of top-performing stations. For the purposes of this project, 100 top-performing stations will continue to be used to extract insights. The follow-up questions section will discuss ideas for future quantitative analysis that could be performed if this number of stations needs to be filtered (reduced) from 100.

🔆 Key Insights

The 100 top-performing stations achieved a 42% market penetration in terms of bike share ridership across the Bike Share Toronto network during 2022.