Business Management Case Study

Course arrangement

The goal of analyzing business data is to make business data-driven decisions, and we usually make use of a series of analysis methods during the decision-making process. At first, the explorative method is needed to be applied with data visualization to explore the implicit structure and trends and to understand the crucial problems. Then, we would apply the diagnostic method to find the casual relationship, apply the predictive method to predict the situation, and make the optimum decision and strategies through simulation and optimization tools. In the business digital transformation process, we have to repeat the mentioned steps to adjust the strategies, as well as digital enterprises. Hence, after introducing the analysis methods and analysis tools, we will design a series of business analytics cases and let students have chances to apply the several analysis methods and analysis tools to make strategies in the last course of each three curriculums.

In the business analytics case study, we plan to refer to the online course, Analytics Edge, offered by MIT to help students understand the business analytics used in different fields, such as sports, medical, police, legal affairs, insurance, and dating sites. Besides, we plan to refer to a series of online courses, from Business Analytics Specialization, which is designed by Wharton. Students will have a chance to enhance the managerial ability by using data in managerial-related cases. Furthermore, we also plan to imitate the “Harvard simulation cases”, using interactive web pages to simulate market segmentation, differential pricing, product design, product line planning, product differentiation, and price competition, etc. Therefore, students can experience the situation when decisions need to be made under time pressure.

Course objective

The objective of the course is to assist students to use exploratory, diagnostic and predictive analysis methods under various business situations. Also, the course provides students with simulation and optimization tools to analyze, predict, and simulate situations.


PART-I Company internal management cases

  • Customer value management
  • Strategic human resources management
  • Managerial accounting
  • Operation (supply chain) management

PART-II Operation and management cases

  • Professional sports manager
  • Medical insurance company
  • New product sales estimation
  • Airline company revenue management
  • Kaohsiung real estate registration
  • Diabetes medical prevention
  • Customer value management in the retail industry
    • Acquire Valued Shoppers Challenge (Kaggle)
    • 349,655,789 transactional items,26,500,000 transaction records,311,500 customers
  • E-commerce website recommendation system
    • Instacart Market Basket Analysis (Kaggle)
    • 22,026,608 transactional items, 2,178,586 transaction records, 131,209 customers, 49,688 products
  • The estimation of the efficiency of discount coupons in the retail industry
    • Coupon Purchase Prediction (Kaggle)
    • 2,833,178 website clicks and 1,046,668 times website views
    • 168,996 transaction records, 22,873 customers
  • E-commerce website browsing record
    • Yoochoose RecSys 2015 Competition Dataset
    • 33,003,944 clicks, 9,297,691 sessions

PART-III Management

  • Data: Yelp Academy Challenge Round 12
    • 5,996,996 review articles (2005~2018),
    • 280,992 photos
    • 1,518,169 reviewers
    • 188,583 stores
  • Analysis steps:
    • Use big data analysis platform to organize structured and unstructured data
    • Use exploratory analysis methods and data visualization tools to see the structure and trends hidden in the data
    • Propose key questions from the structure and trends
    • Use predictive and diagnostic methods, and simulation and optimization tools to propose feasible countermeasures to crucial problems
    • Prepare for final presentation
  • Possible analysis items:
    • Business category analysis
    • Comment content analysis
    • Comment text effect analysis
    • Image (picture) effect analysis
    • Reviewer network analysis
    • Social media content analysis

Teaching materials

Textbooks are not needed. We take previous self-made materials as the basis and renew the contents with the needs of the curriculum. The teaching materials include:

  • Course website: integer the self-made materials and the online resources to facilitate collaborative learning. (HTML)
  • Guided videos: assist students to understand and preview the online contents.
  • Handouts: every unit has its own slides. (PPT)
  • Course notes: code used in every course will be provided. (R-Notebook)
  • Assignment notes: the assignment of every unit will be provided. (R-Notebook)
  • Online simulation: assist students to understand complicated concepts through the interactive simulation. (R:Shiny)
  • Videos: the videos of every course will be recorded and provided for students to review. (YouTube)

Please review the examples via the web links above.

Big data computing resources of NSYSU

The course will use a huge data set of more than 300 million data. Students can log in College of Management: Big data and business analytics platform by student ID and acquire the following resources:

  • All the large datasets will be installed in the Hadoop system in advance.
  • Big data can be loaded into the Spark, and the data can be processed through the data frame interface.
  • Big data can also be loaded into the GreenPlum, and processed through the SQL interface.
  • In addition to the data frame and SQL interface, the integrated analysis platform also provides uploading and exploration functions.
  • Through the platform, a high-end interface (Keras) can be used to create artificial intelligence models using a GPU-equipped deep computing host
  • When the user logs in for the first time, he/she can use the above resources through the demo program under the table of contents.
  • Users can upload and analyze corpus data in a semi-automated manner through the text analysis platform. At present, the text platform has stored text data on most pages of the PTT website in the past 10 years. Users can directly use keywords and dates to filter text content.

Online courses

Other online materials (MOOC’s):

  • A series of Analytics Edge online courses offered by MIT
  • A series of Business Analytics Specialization online courses offered by Wharton
    • Customer Analytics
    • Operations Analytics
    • People Analytics
    • Accounting Analytics

Several units, assignments, or cases are used as preview materials, references, and web-based course materials.

Opening data source

To manage the cases, the course collects data from the following websites:

  • Kaggle
  • UCI Machine Learning Repository
  • Government open data
  • National Center for High-performance Computing

R language special packages

In addition to R and RStudio, the following packages will be used as well:

  • Interactive web graphics package (htmlwidget, …)
  • Interactive simulation package (shiny, shinyWidgets, shinydashboard)
  • Image and data processing package (sf, tmap)
  • Product recommendation system package (recommaenderlab)
  • Revenue management package (fPortfolio, quantmod)
  • Investment portfolio management package (fPortfolio, quantmod)
  • Non-linear programming (optimization) package (modopt.matlab)

Other online resources

We also plan to use the following online services in a free trial during the course:

  • Set up a deep learning server on Google Cloud Platform (GCP)
  • Practice the usage of automated machine learning systems through IBM Watson Analytics
  • Practice operating social network monitoring and text content analysis through IBM Watson Analytics for Social Media

Reference included

  1. Albright, S., & Winston, W. (2014). Business analytics: Data analysis & decision making. Nelson Education.
  2. Edwards, M. R. & Edwards, K. (2016). Predictive HR Analytics: Mastering the HR Metric. Kogan Page.
  3. Mizik, N. & Hanssens, D. M. (2018). Handbook of Marketing Analytics: Methods and Applications in Marketing Management, Public Policy, and Litigation Support. Edward Elgar Pub.
  4. Venkatesan, R., Farris, P., & Wilcox, R. T. (2015). Cutting-edge marketing analytics: real world cases and data sets for hands-on learning. Pearson Education.