1
Engineering
Data Analysis
INTRODUCTION
2
Introduction
We live in an age of technology… has two important implications:
Has made it possible to collect large amounts of data
- It has become relatively easy to collect data.
- Point-of sale data, credit data, stock data, economic data etc..
- As a result, data is plentiful.
- Challenge: how to make sense of the collected data.
Technology has made it possible to analyse data and make decision making
based on quantitative analysis.
By using quantitative methods, information in data sets can be uncovered.
Decision making can be made to take actions based on the quantitative analysis.
3
Introduction
Objective of the course
To be exposed to a variety of quantitative methods to analyse data and decision
making.
Provide a hand-on approach to exploring quantitative methods and illustrate their
use in a variety of business situations.
Make use of simple analytic tools such as Excel to get exposure to quantitative
methods for data analytics.
4
Course Overview
Ch1 – Introduction to Business Analytics
Part 1 – Exploring Data
Ch2 – Describing the Distribution of a Simple Variable
Ch3 – Finding Relationships amongst variables
Part 2 – Probability & Decision making under uncertainty
Ch4 – Probability & Probability distribution
Ch5 – Normal, Binomial, Poisson and Exponential Distributions
5
Course Overview
Part 3 – Statistical Inference
Ch7 – Sampling and Sampling Distributions
Ch8 – Confidence Interval Estimation
Part 4 – Regression Analysis & Time Series Forecasting
Ch10 – Regression Analysis: Estimating Relationships
Title: Business Analytics: Data Analysis &
Decision Making
Prescribed Text Authors: S. Christian Albright & Wayne L.
Winston
Book
Edition: 6th Edition
ISBN-13: 978-1-305-94754-2
6
7
30% for Class Assignments ( x 5)
Evaluation
Criteria 30% for Semester Project
40% for Final Evaluation
Subminimum Requirement (40%)
8
Introduction to Modelling & Models
A model is an abstraction of a Tries to capture the essence and key features
of the problem
real problem.
Graphical Models
Three types of models Algebraic Models
considered in this course: Spreadsheet models
9
Graphical Models
Attempt to portray how different elements of a problem are
related highlighting what affects what graphically.
They do not provide quantitative details on how to solve the
problem.
The purpose is to show the important elements of a problem
and how they are related.
10
Graphical Models
Figure 1.3: Graphical Model of an Ordering Decision
11
Algebraic Models
Algebraic models use algebraic equations and inequalities to specify a
set of relationships in a very precise way.
• Can be stated concisely and with great generality.
Product mix problem
• A company can make several products.
• Each of which contributes to a certain amount of profit, and each consumes a
certain amount of scarce resources.
• Problem is to select the product mix that maximises profit, subject to limited
availability of the resources.
12
Algebraic Models
Algebraic Product Mix Model • xj is the amount of product j
• uj is an upper limit on the amount of
max 𝑝𝑥
product j that can be produced
Subject to • pj is the unit profit margin for product j
• aij is the amount of resource i
𝑎 𝑥 𝑏 ,1 𝑖 𝑚 consumed by each unit of product j
• bj is the amount of resource i
0 𝑥 𝑢 ,1 𝑗 𝑛 available
• n is the number of products
• m is the number of scarce resources.
13
Spreadsheet Models
Use a ‘spreadsheet’ to relate various quantities.
Has to be well designed and well documented.
Provides an intuitive way of relating various quantities.
14
A Seven-Step Modelling Process
1. Define the problem – identify the underlying problem
2. Collect & Summarize data – gather the right data and summarise the data appropriately.
3. Develop a model – the form of the model varies from problem to problem, but must
capture the important elements of the problem.
4. Verify the model – determine whether the developed model is an accurate representation
of reality.
5. Select one or more decisions – Once the model has been verified, appropriate decisions
can be made.
6. Present the results – demonstrate the benefits of results obtained from a model
7. Implement the model and update it over time.
15
Summary
Decision making objectives can be numerous: prediction,
forecasting, product mixes, selection, classification,
optimisation etc.
The question is how can better decisions be made by
understanding data better.
The objective of this course is to provide you with the skills
required to analyse data that provide the foundation for
making better decisions.
