A factory is starting a project to reduce unplanned downtime. A new analyst wants to choose a model family before anything else is written down. Where should this life cycle begin?
Select an answer to reveal the explanation.
Short Explanation
You do not pick the truck before you know what has to be delivered. The life cycle starts with the business question, and the model family comes after that question is clear. The trap is treating the algorithm list as step one.
Full Explanation
A modeling life cycle starts from the business question: which downtime events matter, who will act, and what decision a useful result would change. Choosing a model family is a later design step that depends on the outcome, the data, and the cost of a wrong call. Locking an algorithm on day one skips the problem the factory actually hired the team to solve. A shortlist of model families before anyone inspects the records or agrees how a score would be used treats the algorithm as the project. A live scoring endpoint is a deployment concern that belongs after a model exists and after the decision is defined. A hyperparameter grid for a preselected algorithm assumes both the goal and the learner are already settled, which they are not at the start. Exam caveat: business understanding can be revisited later, but that loop does not make model selection a valid first step. Operational check: read the project brief and confirm it names the downtime decision and the user of the result before any algorithm is chosen.