Advanced MBA assignment on CRISP-DM, credit risk modelling and business intelligence dashboards helps you practice real-world credit scoring, BI implementation and sales performance analytics using RapidMiner and Tableau in a global business context.

Assignment alignment and learning goals

Assignment 4 relates to the specific course learning objectives 1, 2 and 4 and associated MBA program learning goals and skills: Global Content, Problem solving, Change, Critical thinking, and Written Communication at level 3. In many contemporary MBA curricula, these capabilities are explicitly linked to evidence-based analytics and data-driven decision making in volatile global markets, especially around risk and performance management in financial services and other data-rich industries.

1. demonstrate applied knowledge of people, markets, finances, technology and management in a global context of business intelligence practice (data warehouse design, data mining process, data visualisation and performance management) and resulting organisational change and how these apply to implementation of business intelligence in organisation systems and business processes. As organisations accelerate digital transformation, managers are expected to integrate BI platforms, predictive models and dashboards into core processes to support agile responses to customer behaviour, regulatory change and competitive pressure.

2. identify and solve complex organisational problems creatively and practically through the use of business intelligence and critically reflect on how evidence based decision making and sustainable business performance management can effectively addressing real world problems. Recent research in credit analytics and BI shows that combining structured data mining methods such as CRISP-DM with performance dashboards enables more transparent, auditable and sustainable decision processes across functions.

4. demonstrate the ability to communicate effectively in a clear and concise manner in written report style for senior management with correct and appropriate acknowledgment of main ideas presented and discussed. Senior leaders increasingly expect concise, insight-rich reports complemented by clear visualisations, making written communication and dashboard storytelling a core differentiator for data-savvy MBA graduates.

The key frameworks, concepts and activities covered in modules 2–12 and more specifically modules 6 to 12 are particularly relevant for this assignment. These modules typically cover the full CRISP-DM lifecycle, predictive modelling for credit risk, BI system implementation issues and practical dashboarding for performance management using tools such as RapidMiner and Tableau.

Overall assignment description

This assignment consists of three tasks 1, 2 and 3 and builds on the research and analysis you conducted in Assignment 2. In many BI and analytics courses, scaffolding assignments in this way helps you deepen your earlier exploratory work into more robust modelling, evaluation and communication suitable for decision-makers.

Task 1 is concerned with developing and evaluating a model of key factors impacting on credit risk ratings for loan applications in determining whether approve a loan or not approve a loan. Contemporary credit scoring practice uses similar supervised learning approaches to segment applicants, estimate probability of default and support consistent accept/reject and pricing decisions under regulatory constraints.

Task 2 is concerned with the key opportunities and challenges associated with the implementation and utilisation of business intelligence systems. Current literature highlights issues such as data governance, user adoption, integration with legacy systems and the need for continuous model monitoring when embedding BI and analytics into organisational routines.

Task 3 is concerned with performance management and provides you with the opportunity to design and build a sales performance dashboard using pivot tables and Tableau 7.0 Desktop. Modern Tableau banking and credit dashboards typically combine customer segments, risk metrics and time-based performance indicators to support proactive portfolio and sales management.

Task 1: CRISP-DM and credit risk modelling (40 marks)

In Task 1 of this Assignment 4 you are required to follow the six step CRISP DM process and make use of the data mining tool RapidMiner to analyse and report on the creditrisk_train.csv and creditrisk_score.csv data sets provided for Assignment 4. The CRISP-DM framework remains a widely adopted standard in credit analytics projects because it structures projects from business understanding through deployment, and tools such as RapidMiner provide built-in operators and templates that map directly onto these phases.

You should refer to the data dictionary for creditrisk_train.csv (see Table 1 below). Up-to-date credit scoring case studies recommend using the data dictionary to guide feature selection, identify leaky or redundant variables and document assumptions for regulators and internal stakeholders.

In Task 1 and 2 of Assignment 4 you are required to consider all of the business understanding, data understanding, data preparation, modelling, evaluation and deployment phases of the CRISP DM process. Recent empirical work on credit classification using CRISP-DM stresses that iteration between these phases, particularly between modelling and evaluation, is often needed to reach a robust, generalisable model that aligns with business constraints.

Task 1a: Business understanding and variable selection

a) Research the concepts of credit risk and credit scoring in determining whether a financial institution should lend at an appropriate level of risk or not lend to a loan application. Contemporary sources describe credit scoring as a quantitative tool that estimates the likelihood of default or delinquency, often combining behavioural, demographic and financial variables in a scorecard or machine learning model.

This will provide you with a business understanding of the dataset you will be analysing in Assignment 4. Many up-to-date guides emphasise starting from the business problem, including risk policy, target portfolio mix and regulatory requirements, before finalising target variables and performance metrics for the credit model.

Identify which (variables) attributes can be omitted from your credit risk data mining model and why. Recent credit risk modelling practice recommends omitting variables that are identifiers, data collection artefacts, post-outcome information or proxies for protected characteristics, and using exploratory analysis plus domain knowledge to justify exclusions.

Comment on your findings in relation to determining the credit risk of loan applicants. Linking your selected predictors to known drivers of default such as income stability, indebtedness, prior delinquencies and loan-to-value ratios helps demonstrate that your model is both statistically sound and grounded in established credit risk theory.

Task 1b: Exploratory analysis and predictors

b) Conduct an exploratory analysis of the creditrisk_train.csv data set. Recent EDA practice in credit scoring includes profiling variables, visualising distributions, checking correlations and using univariate and bivariate analyses to assess predictor strength before full model building.

Are there any missing values, variables with unusal patterns? Current literature suggests quantifying missingness patterns, assessing whether data are missing at random and deciding between imputation, binning or exclusion to avoid biased risk estimates.

How consistent are the characteristics of the creditrisk_train.csv and creditrisk_score.csv datasets? Modern CRISP-DM applications recommend checking for dataset shift between training and scoring data, since substantial distribution differences may reduce model performance and require recalibration or segmentation strategies.

Are there any interesting relationships between the potential predictor variables and your target variable credit risk? (Hint: identify the variables that will allow you to split the data set into subgroups). Using decision trees, weight-of-evidence plots or partial dependence charts can reveal meaningful splits and non-linear effects that guide both model choice and business interpretation of risk drivers.

Comment on what variables in the data set creditrisk_train.csv might influence … In applied research, such commentary often highlights variables like credit history, utilisation ratios, income levels, employment status and collateral value as key factors influencing predicted default or downgrade risk across portfolios.

A strong sample answer for Task 1 would briefly define credit risk as the likelihood that a borrower fails to meet contractual payment obligations, then explain how credit scoring models translate applicant and account variables into a probability of default used for accept/reject and pricing decisions. It would then describe how the CRISP-DM phases were applied in RapidMiner, from clarifying the business objective with stakeholders, through cleaning and transforming the creditrisk_train.csv data, to training and evaluating a suitable classification model such as a decision tree or logistic regression. The answer would justify omission of non-informative or inappropriate variables using both EDA evidence and ethical/regulatory considerations, and discuss how key predictors like prior delinquencies and debt-to-income ratios shaped risk differentiation across subgroups. Finally, it would reflect on how the resulting model and its performance metrics could be deployed within a BI environment to support transparent, data-driven lending decisions and ongoing portfolio monitoring for senior management.

---

References

Altair Engineering Inc. (2025) ‘Credit Scoring Series Part Three: Data Preparation and Exploratory Data Analysis’, Altair Blog, 28 October. Available at: https://altair.com/blog/articles/credit-scoring-series-part-three-data-preparation-and-exploratory-data-analysis.

Altair Engineering Inc. (2025) ‘Credit Scoring: Credit Scorecard Modeling Methodology’, Altair Blog, 26 August. Available at: https://altair.com/blog/articles/credit-scoring-series-part-two-credit-scorecard-modeling-methodology.

Siregar, R., Sari, R. and Sari, D. (2020) ‘Credit Classification Using CRISP-DM Method on Bank XYZ’, International Journal of Emerging Trends in Engineering Research, 8(6), pp. 2553–2559. Available at: http://www.warse.org/IJETER/static/pdf/file/ijeter28862020.pdf.

Printalect (2023) A Home Credit Risk Analysis & Broad Application of CRISP-DM. Available at: https://printalect.github.io/Credit-Risk-Analysis-FV3/Home-Credit-Risk-Analysis-Report.pdf.

Tableau Software (2024) ‘How Financial Services Can Reduce Credit Risk with Data-Driven Analytics’, On-Demand Webinar, 7 February. Available at: https://www.tableau.com/learn/webinars/financial-services-dashboard-series-credit-risk.