Uber Supply-Demand Gap Analysis & Possible Solutions
Abstract
Aim:
The objective is to identify and analyze the problems in Supply-Demand of Uber trips from City to Airport and vice-versa across various hours of the day and find out possible solutions to address the problem.
Problem Statement:
Uber is an app-based transportation network and taxi company. In its Airport rides in a particular city, many of its users face the problem of cancellation by the driver or non-availability of cars.
These very issues impact the business of Uber and it loses out on its revenue.
Approach:
The primary approach is to use EDA operations like univariate and segmented univariate analysis to identify problems and tackle it.
Using the data set available, we make use of R, a statistical computing program to perform various EDA (Exploratory Data Analysis) operations and hence find out the results. We make use of ggplot2 package of R to plot graphs.
Inference:
To come up with solutions to address this Supply-Demand issue and recommend realistic, actionable solutions.
Problem Solving Methodology
Initial Analysis: Understand the dataset given. Look through its structure, identify the datatypes of various columns and get a basic idea of the dataset to proceed further.
Data Cleaning: Look out for visible data quality issues and rectify them. Check for blanks, duplicate data and convert certain columns to required datatypes.
Exploratory Data Analysis: Making use of R, carry out various EDA operations like Univariate and Segmented Univariate analysis and come up with intuitive insights of the Supply-Demand problem.
Visualization: Make use of ggplot2 package to plot various graphs with proper aesthetics and geometry, clearly displaying important insights.
Analysis
1. In the this plot, we see that the time slots between 5.00 to 9.00 and 17.00 to 21.00 are particularly busy with more demand.
2. It is also seen that between 5.00 to 9.00 time slot, there are more number of cancelled cabs.
3. Between 17.00 to 21.00 time slot, there are more number of “No Cars Available”
1. From the this plot, it is clear that between 5.00 to 9.00 time slot, there are more number of requests in the city.
2. Between 17.00 to 21.00 time slot, there are more number of
requests from the Airport.
Analysis — Time Slots
1.To carry out further analysis, we create time slots based on the number of trips.
2.From our observations from previous plots, we notice that the times between 5.00 to 9.00 and 17.00 to 21.00 have high request and hence are of our interest.
3.We can make various time slots like:
- 0.00 to 5.00 — Early Morning
- 5.00 to 10.00 — Peak Morning
- 10.00 to 17.00 — Normal Day Time
- 17.00 to 22.00 — Peak Evening
- 22.00 to 0.00 — Late Night
4. From the plot it is confirmed that in the Peak_Morning time slot, there are more number of cancelled cabs.
5. In the Peak_Evening time slot, there are more number of “No Cars Available”.
Analysis — Peak_Morning Time Slot & Cancellations in the City
The above plot makes it clear that in the Peak_Morning, between City and Airport, the City has maximum Cancellation of trips.
This Pie Chart indeed confirms that there are high number of cancellation of trips in the City
Analysis — Demand, Supply and Gap in the City
1. We assume that:
▪ Demand = Total No. of Requests
▪ Supply = Total Number of Completed Trips
▪ Gap = Demand — Supply
▪ “Cancelled” and “No Cars Available” are Gaps
▪ “Trip Completed” is Supply.
2. We observe:
▪ Demand or Total Requests = 1678
▪ Supply or Trips Completed = 472
▪ Gap = 1205
Clearly, we see a big Gap in supply from the plot.
Analysis- Peak_Evening Slot & “No Cars Available” in Airport
Peak_Evening Time Slot
The above plot makes it clear that in the Peak_Evening, between City and Airport, the Airport has maximum “No Cars Available”.
No Cars Available in Pickup Point Airport
This Pie Chart indeed confirms that there are high number of “No Cars Available” from the Airport.
Analysis — Demand, Supply and Gap in the Airport
1. We assume that:
- Demand = Total No. of Requests
- Supply = Total Number of Completed Trips
- Gap = Demand — Supply
- “Cancelled” and “No Cars Available” are Gaps
- “Trip Completed” is Supply.
2. We observe:
- Demand or Total Requests = 1800
- Supply or Trips Completed = 373
- Gap = 1427
Clearly, we see a big Gap in supply from the plot.
The Reason for the Issue of the Supply-Demand Gap
Peak Morning Slot Problem
1.We observed that in the Peak Morning Slot where the demand for City to Airport trips are high, there are maximum number of trip cancellations leading to great Gap in the Demand & Supply.
2.Out of the Demand or Total Requests of 1678, the Supply or Trips Completed were only 472, with high Gap of 1205.
3.This is because the trip to Airport from the City usually takes a longer time. Once the driver is in the Airport, he will have a longer idle time depending on the on the flights patterns. In the mornings, a lot of flights usually leave the city and less flights arrive. Also, it doesn’t make any economical sense to come back empty from the Airport to the City.
4.All these factors lead to high cancellation rate for trips from City to Airport.
Peak Evening Slot Problem
1.We observed that in the Peak Evening Slot where the demand for Airport to City trips are high, there are maximum number of “No Cars Available” leading to great Gap in the Demand & Supply.
2.Out of the Demand or Total Requests of 1800, the Supply or Trips Completed were only 373, with high Gap of 1427.
3.This is because by the night, a lot of flights including the international ones start arriving at the airport. This creates a high demand for the cars. Also, the Uber driver partners start retiring for the day as the dawn proceeds leading to a high non availability of the cars.
4.These factors lead to high “No Cars Available” issue for trips from Airport to City.
Recommendations to Resolve the Supply-Demand gap.
1.Uber should provide higher monetary incentives for driver taking a Peak Morning trip from City to Airport.
2.Setting targets for drivers to take up the City to Airport or Airport to City trips and then suitably rewarding them through lucky draw schemes, gift vouchers.
3.Incentivize Airport to City trips in the night so that more drivers from the City reach the Airport thus increasing the availability of the cars.
4.Ask its drivers to not retire from services early in the evening.
5.Conduct orientation classes for drivers and let them know about the insights obtained analysis of this problem so that the drivers know where to be at what time of the day.
