ANLY482 AY2017-18T2 Group08 : Project Overview

Homepage

Our Team

Project Overview

Project Findings

Project Management

Documentation

ANLY482 AY2017-18 T2 Projects

Description

Methodology

Project Background

in progress

Project Motivation

Sponsor Motivation

Currently, oBike has engaged two full-time employees as well as third-party contractor to manage these illegal parking cases. When issued with a new ticket, it is the job of these employees to record the case details into a spreadsheet. Due to the overwhelming number of illegal parking cases occurring each day, oBike has not yet been able to take a closer look at their data. As such, they have agreed to collaborate with us in the hopes of using analytics and its related tools to clean, analyse and make sense of the data that they have. In doing so, they hope to be able to better manage their illegal parking cases.

Team Motivation

As Albert Einstein had once said, “Education is what remains after one has forgotten what one has learnt in school.” Keeping in mind this mantra, we understand the importance of applying what we have learnt in the world of business. Further, as final-year students with a second major in Analytics, this module serves as a platform for us to develop our analytical skills and to gain legitimate hands-on experience. It gives us a taste of what analytics has to offer, thereby preparing us for the real world.

In today’s world, environmental degradation, climate change and sustainability issues pose a great problem to businesses and individuals alike. oBike’s mission statement is to build a sustainable mode of transportation for public masses and to achieve energy savings and reduce carbon dioxide emissions globally. As we strongly resonate with their mission and values, we have decided to collaborate with oBike to value-add to their business via analytics. The scope of this project also challenges us to take the initiative to acquire greater knowledge and develop new skills above and beyond what we have learnt in school. As individuals with a zest to learn more, this project will definitely be an eye-opening experience.

Project Objective

General Objectives

Keeping in mind oBike’s current business problem, this practicum ultimately seeks to help them tackle it via the following objectives:

(i) Identify hotspots for illegal parking cases

(ii) Project the illegal parking patterns by analysing historical data

(iii) Determine suitable areas for yellow boxes to be painted

Intermediary Objectives

Having reviewed the sample data provided by our sponsor, we arranged a consultation with our supervisor. Following that, our team has come up with additional interim goals to ensure that the aforementioned objectives are achievable. These include:

(i) Developing a better understanding of oBike’s third-party contractor’s operations

(ii) Perform exploratory data analysis to obtain a rough understanding draw a summary of the data

(iii) Properly define terminologies i.e. what constitutes ‘high-risk’ by determining a suitable criteria (e.g. fines issued more than x times per day)

This project scope is focused entirely on Singapore, due to limitations in the data available.

Scope of Work

The following section will discuss the scope of work we hope to cover during the course of this practicum. However, as we are still in the infancy stages of this collaboration, the true scope of work and methods to be used may (or will be) subject to changes.

Project Discovery

To begin with, we developed an understanding of oBike’s business domain so as to discover business goals, requirements and problems the client is currently facing. This was done via a face-to-face meeting with a client where we had an initial understanding of the illegal bike parking problem that they faced, as well as the hefty fines issued by LTA. In addition, secondary research was done to better understand oBike's competitors, LTA regulations as well as the bike-sharing industry as a whole.

Data Preparation

Data has been collected from oBike in the form of CSV files. Following which, data preparation and cleaning will be carried out. After, exploratory data analysis (EDA) will be conducted to develop a better understanding of the data set. Summary statistics will enable us to get a higher-level view of the data while dashboards will be created to help us to visualize characteristics of the data.

Model Planning and Building

Results from the EDA phase shall then be used for model planning and development. The model should be validated before being used for analysis and the generation of insightful findings for oBike. We plan to use a 3-step process to fulfil the client's objectives as follows:

(i) Descriptive analytics – Use of data mining techniques such as geospatial analysis and data visualisation to develop an in-depth understanding of current illegal bike parking patterns

(ii) Predictive analytics – Use of statistical models and forecasting techniques such as time series forecasting to predict future illegal bike parking trends, for development of a preventive approach.

(iii) Prescriptive analytics – Use of existing data and forecasted trends to suggest optimal locations for additional yellow boxes to be painted by bike-sharing companies to reduce illegal parking problems.

Communication of Insights

Throughout the duration of this project, the project progress will be communicated to the client via weekly updates. Finalised insights obtained from the model would then be communicated to the client via a presentation. These findings will also be shared via an analytics practicum conference and submitted in the form of a final paper.