AI for Operations Capstone Exercise — AI for Operations Managers

What you'll learn

Apply skills from across this course in a single realistic professional scenario
Produce a concrete, role-relevant deliverable using AI tools
Self-assess your output against professional quality criteria

Operations leaders who want to introduce AI into their organisations rarely face a purely technical challenge. The harder work is the business case: quantifying the problem clearly, proposing a solution with credible specifics, and directly addressing the organisational history that makes sceptics sceptical. This capstone puts you in that position.

You are an operations director at a third-party logistics company. Two AI pilots have already failed at your organisation. The executive team is willing to hear a proposal for AI-assisted demand forecasting, but they want a rigorous case that explains why this one is different. You will use AI to accelerate the drafting of each section of that business case while exercising the domain judgment that no AI tool can supply.

Capstone Exercise

Build the Business Case for AI-Assisted Demand Forecasting

Context

You are the operations director at a third-party logistics firm with 12 distribution centres and a peak-season forecast error rate consistently running above 18%. Two previous AI pilots failed: the first collapsed due to poor data quality in the ERP, the second due to inadequate change management and low adoption on the warehouse floor. The executive team has agreed to one more proposal. They expect you to address the previous failures directly, not treat them as irrelevant history.

Your Task

Use Claude or ChatGPT to produce a first draft of each of the following five sections of a business case document: (1) Current state problem quantification, including specific operational costs tied to the forecast error rate; (2) Proposed AI solution with a named capability description, such as probabilistic demand forecasting integrated with your existing ERP; (3) Change management section that directly addresses why the two previous pilots failed and what is structurally different this time; (4) A risk register with at least four risks and a mitigation for each; (5) A success metrics framework with three to five measurable KPIs. For each section, write one annotation directly below the AI draft that identifies the specific human judgment call, domain knowledge, or internal data point that the AI could not supply and that you would need to verify or override before the section is ready to present.

Your notes (optional)

Deliverable

A five-section business case document of 600 to 750 words with one annotation per section identifying the human judgment or domain verification requirement. The document demonstrates where AI accelerates operations analysis and where the operations director's expertise is irreplaceable.

Quick check

The change management section has to account for two failed pilots, and the capstone asks you to prompt for a section that addresses each one structurally rather than diplomatically. What is the difference it is drawing?

Select one answer.

Key takeaways

AI can compress the first-draft time on structured analytical documents significantly, but business case credibility depends on the human judgment layer: real operational data, honest assessment of previous failures, and domain-specific risk identification.
The change management section of any AI proposal is where most proposals succeed or fail with sceptical executive audiences: directly addressing what went wrong previously is more persuasive than ignoring it.
A risk register with specific mitigations tied to your operational context is more valuable than a generic one, and the annotation exercise reveals which risks the AI identified from pattern-matching versus which only an operational insider would know to include.
The annotation discipline practised in this capstone is a professional habit worth keeping: whenever AI drafts a document that will influence a real decision, mark the sections that require human verification before that document leaves your hands.