Coding and Automating Decision Intelligence is a hands-on, 24-hour training program (delivered  in 8 x3hr sessions) designed for business analysts, data professionals, and strategic decision-makers who want to supercharge their modeling capabilities. Whether you're managing uncertainty in supply chains, forecasting volatile markets, or guiding investment decisions, this program equips you to build automated, simulation-powered decision models using SQL, Power BI, and Julia. 
 
You'll learn how to consolidate data from multiple sources, run probabilistic simulations, and publish risk dashboards that deliver insights at the speed of business. By the end, you'll have the tools—and confidence—to transform your decision models into repeatable, scalable intelligence engines.
 
Module 1: Foundations of Decision Modeling & Data Wrangling
Objective: Introduce decision model frameworks, uncertainty modeling, and foundational data skills.
Principles of probabilistic modeling and decision support
Identifying input-output relationships and model fidelity
Intro to structured data: rows, tables, variables
SQL crash course (SELECT, WHERE, JOINs)
Lab: Write your first SQL queries on simulation-ready data
Tooling overview: Excel, Power BI, Julia, and their roles
 
Module 2: Modeling Uncertainty with Distributions
Objective: Teach analysts to define realistic input behaviors using probability distributions.
When and why to use probabilistic inputs
Bounded vs unbounded distributions
Fitting distributions with historical data
Choosing and validating distributions
Lab: Fit, compare, and simulate distributions using Julia and MCHammer.jl
Integrating expert opinion into models
 
Module 3: Modeling Relationships with Correlation & Copulas
Objective: Teach how to link uncertain variables to reflect real-world behavior.
Correlation vs causation: theory and application
Rank-based correlation and Spearman's method
Advanced dependency modeling with Copulas
Lab: Build a correlation matrix and simulate using corvar() in Julia
Real-life cautionary tales (e.g., mortgage crash)
 
Module 4: Forecasting & Time-Series Simulation
Objective: Provide tools for predictive modeling using time-series and Monte Carlo logic.
Time-series components and uncertainty over time
Log returns, exponential smoothing, ARIMA
Corridor forecasts (confidence bands)
Lab: Build a corridor forecast in Julia from raw historical data
Model deprecation and model maintenance best practices
 
Module 5: Building Simulation Models in Julia
Objective: Move from theoretical to practical simulation modeling.
Writing simulation loops in Julia
Sampling techniques and vectorized simulations
Defining inputs, scenarios, and outputs
Lab: Build a Monte Carlo model with inputs, outputs, and KPIs
Debugging models using sensitivity analysis
 
Module 6: Optimization under Uncertainty
Objective: Automate decision selection using optimization algorithms under probabilistic inputs.
Project and portfolio optimization
Efficient frontier and viability-fit methods
Combining simulation with optimization (Monte-Carlo optimization)
Lab: Build a portfolio optimization model using Julia + simulated inputs
Using objects vs arrays for model performance
 
Module 7: Automating Data Refresh and Reporting with SQL & Power BI
Objective: Enable repeatable execution of models and dynamic dashboard generation.
Creating and updating dashboards with Power BI
Connecting Power BI to simulation outputs
Automating SQL pipelines to fetch model data
Lab: Create a Power BI dashboard for a simulated supply chain model
Discuss security, governance, and sharing best practices
 
Module 8: Integrated Case Study – From Model to Dashboard
Objective: Apply end-to-end workflow on a real-world problem.
Choose one of the following cases:
Building vs outsourcing: global manufacturing strategy
Capital allocation under economic uncertainty
Sales forecasting for product launch
Supply chain resilience strategy
Build full model (SQL + Julia)
Simulate scenarios & optimize
Publish live dashboard
Peer review & feedback