Full document

`summarise()` regrouping output by 'week' (override with `.groups` argument)

• First and last weeks are partial
• No decline over the program

`summarise()` ungrouping output (override with `.groups` argument)

• Steady increase to peak on Saturday
• Pattern likely mirrors businees revenue

`summarise()` regrouping output by 'hour' (override with `.groups` argument)

• Spikes everyday at lunch & dinner
• A lot more lunch outings on Friday
• Is that a Monday happy hour peak?

Average paid meter fee:

[1] 1.86

Total meter revenue:

[1] 51442.19

Estimated yearly revenue:

`summarise()` ungrouping output (override with `.groups` argument)

[1] 260640.1

`summarise()` regrouping output by 'spacename', 'metertype', 'meter_lat' (override with `.groups` argument)

• Multispot bias

• Average 5.6 spaces per multimeter

• Balanced usage
• Good design

• Hard to define
• Comparing apples to mystery fruit
• My strategy: interval overlap
  • Computationally intensive
  • Hours of calculations

• City estimates they will take a loss due to secondary costs (vandalism, consultant)

• Why did they cancel early?

  • On track for profit
  • Meters already paid for
  • What report did they see?

• Use SFpark as a template
  • Demand based flexible pricing (Lyft model)
  • Prices were capped
  • Sometimes cap $ wasn’t enough
• Increase price!
  • $1.80 vs $1.50 per hour
  • Who cares about $0.30?
  • 1st hour in garage remains free

• Better anlysis requires more data
  • Business revenues (same period)
  • Parking garage usage/revenues
  • Total cost (consultants, maintanence, repair)
  • Tickets written (same period)
• Call for open data
  • Make ALL government data/reports available
  • Public input should be data driven not emotional
  • Best if data is available real-time
    • So the community can see it first hand
  • Hire data pros to help