Understanding the foundations of total cost management, data for cost engineers and project control professionals, and the role of AI in improving predictability and decision-making.

Provide a brief overview of the fundamentals of TCM, examining the interoperability of the key subdomains and their contributions to project success. Dive into the fundamentals by comparing the various types of data analysis, examining variables, and analyzing the impact of sample size and sample bias.

• Understand the elements of TCM and their importance, specifically cost engineering and project controls as well as planning, measurement, and assessment qualities. Provide the benefits of Total Cost Management that supports an advanced analytics initiative.
• Review of basic statistics for TCM that can contribute to the successful implementation of Machine Learning techniques (key statistical measures and their importance).
• Define the five primary types of data analytics: planning, descriptive, diagnostic, predictive, and prescriptive.

Activity 1: Assess your organization in terms of TCM and AI maturity.

From buzzwords to project delivery relevance – demystifying AI and its role in capital projects.

Organizations and their project teams need a clear, practical understanding of artificial intelligence—not just the algorithms, but how these tools translate into business impact. This module defines foundational AI and machine learning (ML) terms and introduces a framework to link AI capabilities directly to project outcomes.

• Gain a concise definition of artificial intelligence, machine learning, and deep learning.
• Understand the core types of machine learning: supervised, unsupervised, reinforcement, and self-supervised.
• Explore a use-case decision framework that enhances the strategic value of AI in capital projects.
• Discuss real-world examples across the capital project lifecycle where AI adds measurable value.
• Review how these AI functions align with key business drivers: margin, risk, and accountability.

Activity 2: Defining Business Value. Individuals will use a defined strategic approach to assist in problem framing and use-case value.

Beyond automation – how generative intelligence and large language models are reshaping executive decisions.

The rise of Generative AI and LLMs has created transformative opportunities, and risks, for capital-intensive organizations. This module introduces the mechanics and applications of generative AI and provides a foundation for governing AI responsibly in a high-stakes environment.

• Understand what generative AI is, how it works, and where it fits in the AI landscape (including agentic AI and agents).
• Explore the capabilities and limitations of LLMs (e.g., ChatGPT, Gemini, Claude), and learn about risks including bias, hallucinations, privacy, and regulatory gaps.
• Introduce the principles of responsible AI: including fairness, transparency, accountability, safety, and governance.
• See where LLMs can enhance project controls: summarizing reports, developing SOPs, drafting RFIs, etc.

Activity 3: Trade Off of Privacy vs Utility Exercise. Individuals will assess the pros and cons of using GenAI.

Understanding the core techniques for developing relationships of the project data.

Perhaps the single most essential tool in statistical modeling, regression is central to data literacy. Learn what regression does, how it fits into prediction, and what questions to ask data analysts (and yourself).

• Feature selection from project data (cost drivers, schedule delays, change order patterns).
• Define simple and multiple linear regression in cost estimation and performance forecasting.
• Identify independent variables (drivers) and dependent variables (outcomes) in regression models.
• Understand how regression supports scenario modeling and sensitivity analysis in TCM.
• Use case examples: e.g., regression analysis to estimate final cost based on productivity rates and work progress.

Activity 4: "Where do you lose the most predictability in your projects?" Share insights on potential data-driven interventions.

Future-focused cost management using predictive analytics and the basic types of ML and their application to project data.

Understand predictive analysis, the construction of a prediction model, the application of machine learning and AI to prediction, and the strategic implications of prediction.

• Examine the predictive analytics pipeline: define the question → gather data → build a model → validate → apply
• Explore supervised vs. unsupervised learning and their application in project controls
  • Key algorithms: Regression, Decision Trees, Random Forests, Neural Networks
  • When to use what: Forecasting, clustering, classification
• Interpreting model outputs and understanding the practical challenges of the dataset.
• Learn how to communicate prediction results to project stakeholders in non-technical language

Technical Lab: Hands-On Lab (Excel): Simple regression model predicting cost at completion using past project data. Example use cases may include EAC prediction, margin erosion detection, etc.

From data to decisions – designing and testing cost strategies

Molding raw data into a business strategy? Examine cause and effect in data analysis, design an experiment, analyze causal inference without experiments, and be prepared to prescribe a response to the likely results.

• Learn the foundations of experimental design: control groups, variables, and outcome tracking
• Examine how to simulate cause-effect relationships in projects without controlled experiments
• Apply counterfactual thinking to analyze changes (e.g., “What if we had accelerated procurement?”)
• Explore how AI can generate prescriptive options based on predictive model outputs

Defining opportunities for applying large language models for use in the project controls environment

Learn how to harness the power of Large Language Models (LLMs) and AI agents to monitor project health, generate insights, simulate control decisions, and automate responses. Understand the requirements to build tools to improve performance across cost, schedule, and risk.

• Understand the fundamentals of LLMs: context windows, prompting, and response tuning.
• Explore use cases for LLMs in cost engineering, forecasting, risk communication, and change analysis.
• Develop prompt engineering skills to extract structured outputs (e.g., risk registers, trend summaries) from unstructured inputs.
• Understand the use of AI agents that can be used to simulate stakeholder decisions, draft scenario plans, and generate mitigation strategies.

Technical Lab (Hands-On): Create GPT-powered prompts for review risks of a capital project. Feed project data into the model, write structured prompts, simulate a response loop for forecasting overruns, and analyze its recommendations using historical outcomes

Recap of major concepts.

• Recap on major concepts and learning outcomes
• Participant feedback and networking opportunity
• Final Q&A