Careem: A car within 15 minutes
(2013) MIO920Production Management
- Abstract
- Background: Careem FFZ Ltd is a technology startup within the ground
transportation industry based in Dubai, U.A.E, initially
focusing on the chauffeur driven personal transport segment.
In an industry with a fairly low utilization rate and poor
customer experience overall the company aims at setting a
new benchmark for customer service. Being a first mover on
the market with an on-demand smartphone enabled booking
service aims at overcoming many of these deficiencies.
Purpose: A key success factor to a high customer satisfaction and a
broad market penetration is to minimize the lead-time from a
point of booking until the car arrives. The main purpose of
this thesis has been to analyze the problem and establish
models and tools... (More) - Background: Careem FFZ Ltd is a technology startup within the ground
transportation industry based in Dubai, U.A.E, initially
focusing on the chauffeur driven personal transport segment.
In an industry with a fairly low utilization rate and poor
customer experience overall the company aims at setting a
new benchmark for customer service. Being a first mover on
the market with an on-demand smartphone enabled booking
service aims at overcoming many of these deficiencies.
Purpose: A key success factor to a high customer satisfaction and a
broad market penetration is to minimize the lead-time from a
point of booking until the car arrives. The main purpose of
this thesis has been to analyze the problem and establish
models and tools to make sure this goal is fulfilled. Ultimately
this has been an issue of matching demand and supply
efficiently.
Theory: The theoretical framework used has centered on clustering
analysis to analyze the geographical scope of the demand.
Time series models such as autoregressive AR, ARMA and
neural networks models in an attempt to develop prediction
models for the intensity scope of the demand.
Results: The results produced have been encouraging but have also
witnessed of the complexity of the problem. Regarding the
geographical scope the clustering analysis proved a good tool
to pinpoint where the demand appears. Out of the time series
models applied only the AR model produced satisfactorily
results but quite surprisingly a 4-week average model well.
12
By implementation of this prediction model as input to a
separate supply model developed Careem now have a datadriven
model for real-time decision-making. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/3917354
- author
- Olsson, Jörgen and Rigner, Erik
- supervisor
- organization
- course
- MIO920
- year
- 2013
- type
- M1 - University Diploma
- subject
- keywords
- Demand, Supply, On-demand, Real-time, AR, ARMA, Neural Networks, Clustering, Taxi, Transportation.
- other publication id
- 13/5455
- language
- English
- id
- 3917354
- date added to LUP
- 2013-07-03 14:18:46
- date last changed
- 2013-07-03 14:18:46
@misc{3917354, abstract = {{Background: Careem FFZ Ltd is a technology startup within the ground transportation industry based in Dubai, U.A.E, initially focusing on the chauffeur driven personal transport segment. In an industry with a fairly low utilization rate and poor customer experience overall the company aims at setting a new benchmark for customer service. Being a first mover on the market with an on-demand smartphone enabled booking service aims at overcoming many of these deficiencies. Purpose: A key success factor to a high customer satisfaction and a broad market penetration is to minimize the lead-time from a point of booking until the car arrives. The main purpose of this thesis has been to analyze the problem and establish models and tools to make sure this goal is fulfilled. Ultimately this has been an issue of matching demand and supply efficiently. Theory: The theoretical framework used has centered on clustering analysis to analyze the geographical scope of the demand. Time series models such as autoregressive AR, ARMA and neural networks models in an attempt to develop prediction models for the intensity scope of the demand. Results: The results produced have been encouraging but have also witnessed of the complexity of the problem. Regarding the geographical scope the clustering analysis proved a good tool to pinpoint where the demand appears. Out of the time series models applied only the AR model produced satisfactorily results but quite surprisingly a 4-week average model well. 12 By implementation of this prediction model as input to a separate supply model developed Careem now have a datadriven model for real-time decision-making.}}, author = {{Olsson, Jörgen and Rigner, Erik}}, language = {{eng}}, note = {{Student Paper}}, title = {{Careem: A car within 15 minutes}}, year = {{2013}}, }