Business Analytics MSc

This module introduces fundamental statistical concepts and key descriptive modelling techniques in data science, while laying a foundation for the general programming skills required by any top modern business analyst (for example, Python/R).

A range of descriptive modelling concepts will be covered (such as feature engineering, clustering techniques, rule mining, topic modelling and dimensionality reduction) against a background of real world datasets (predominantly based on consumer data).

You will learn not only how to successfully implement foundational descriptive techniques, but also how to evaluate and communicate results in order to make them effective in actual business environments.

This module introduces the fundamental concepts and technologies that are used by modern international businesses to store, fuse, manipulate and visualise mass datasets. 

Key concepts include:

• core database and cloud technologies
• data acquisition and cleansing
• how to manipulate mass datasets (focusing on SQL, Hadoop)
• effective solutions to common data challenges (for example, missing data)
• handling geospatial and open data
• visualisation technologies (Tableau, PowerMap, QGIS, CartoDB)
• web visualisation (HTML5)

All content is based around real-world business examples.

This module covers:

• definitions of entrepreneurship/entrepreneurial activity
• the theoretical perspectives underpinning the study of entrepreneurship
• understanding what shapes the practice of entrepreneurship both in different settings (for example, social entrepreneurship, technology, family business, international entrepreneurship, environmental business, social media) and due to contextual influences (for example, influence of gender, policy)

The emphasis in this module is on formulating (modelling) and solving models with spreadsheets. The topics covered include:

• modelling principles
• optimisation and linear programming
• network models
• introduction to integer programming
• key concepts of probability and uncertainty
• decision theory
• queuing systems
• simulation

The module takes a dual approach covering both the business processes and the quantitative models and techniques necessary for supply chain planning and management. It is divided into three major parts.

1. Supply chain concepts and definitions:
   • Fundamental planning and control concepts for supply chain and operations planning: classification of operational and supply systems
   • Inventory - forms, functions, decisions, models
   • Capacity - definitions and planning
2. Forecasting for supply chain and production management:
   • Planning, scheduling and control approaches: aggregate planning, hierarchical planning and control
   • MRP-based planning and control
   • JIT principles, kanban systems
   • Theory of Constraints (TOC)
   • Enterprise Resource Planning (ERP) systems
3. Supply chain collaboration:
   • Planning and control across the supply chain
   • The bullwhip effect
   • Supply chain collaboration approaches – continuous replenishment
   • Vendor-Managed Inventory (VMI)
   • Collaborative Planning Forecasting and Replenishment (CPFR)

An in-depth look at specialised analytical techniques which present significant opportunities within business environments to extract actionable insights. Applications covered include Recommender Systems (for example, collaborative filtering in business), Text Analytics (linguistic processing, social media analysis), Spatial/Temporal analytics (for example, financial time series), Network analytics (for example, social graph analysis) and High dimensional analytics.

This module explicitly focuses on technologies, planning and managerial issues associated with leading big data projects in business. Key concepts revolve around:

• using data analytics in context (integration of qualitative and quantitative approaches, introduction to survey methods and design)
• the full data lifecycle (including data management and security)
• introduction to organisational scale IT infrastructure
• ethics
• project management 
• presentation skills

This module builds on Foundational Business Analytics covering more advanced predictive models and their motivation within business use cases. Students will establish knowledge of state-of-the-art prediction techniques including SVMs, temporal Nearest Neighbour models, Bayesian methods, Ensembles and Deep Learning.

Practical exercises will be set against a range of real world datasets and time series data. Focusing on the applicability of models to real world problems the module will consider the appropriateness and utility of each method with respect to common ''tricky'' data properties in real world data that lead to under-performing models.

Examples include unbalanced classes, heterogeneous input feature types and detrimentally large number of input features. Within the module methods to unpack the various predictive models to understand why they predict what they do and the utility of this information in various business contexts will be covered.

This module is taught primarily using Python against a background of industrial workflow data modelling environments (for example, SPSS Modeller, Orange) where applicable.

Module content is organised around four themes:

1. More ‘advanced’ forecasting techniques (including more advanced time series and causal models)
2. Inventory modelling (quantity discount models; joint replenishment; reorder point – lot size systems; periodic review models; news vendor model; (S-1, S) model; multi-warehouse situations)
3. Shop floor control: WIP and Little’s law; introduction to operations scheduling and sequencing
4. Introduction to distribution logistics modelling, reverse logistics and closed-loop supply chains

Lecture topics may include digital marketing definition and concept, digital marketing media, digital marketing communication strategy, digital advertising, social media marketing, email marketing, mobile marketing, content marketing, e-commerce vs digital vs internet marketing.

Representing the culmination of the programme, you will design, execute and report a research project based on the analysis of real-world or simulated data. This includes an 8,000-word dissertation, exhibits and data visualisations, and will need to satisfy scholarly objectives consistent with the execution of quality applied research in a business or social context.