A high-precision automotive-component production line is being re-tooled and ramped to a new rate. Eleven scheduled activities -- from process design through equipment procurement, installation, controls, commissioning, qualification and ramp-up -- each carry uncertain durations, and six discrete risk events (equipment breakdowns, supplier slips, workforce shortfalls, quality defects and more) can stretch the timeline further. Tamara runs the whole project network as one Monte-Carlo simulation, so every chart below is driven by the same 60,000-iteration model -- never a hand-drawn illustration.
The headline question for a re-tooling programme is simple: when will the line actually be producing at rate? A single most-likely rollup of the activity durations finishes the project in 535 days -- but that one number hides the shape of what could actually happen.
The histogram below takes the 60,000 simulated completion dates and shows how often each lands in a given month. It is the most honest answer to "when will it be done": not a point, but a spread with a clear centre and a long right tail.
The deterministic plan sits at 535 days, far to the left of the bulk of the distribution. The simulated P50 lands at 683 days and the P90 at 783 days, with a mean of 686. Only 1% of runs finish by the deterministic plan -- the tidy 535-day schedule is very nearly a best case, not a forecast. The right-tail mass past the P90 is exactly the slippage a re-tooling sponsor needs to plan reserves against.
A stochastic Gantt shows, per activity, the P50 finish bar with a P10-P90 whisker. Long whiskers mark the activities whose timing is least certain.
The widest whiskers belong to the procurement, installation and commissioning activities -- exactly the ones that compound into the long right tail of the finish histogram above.
Schedule risk is not spread evenly across the eleven activities. The criticality tornado ranks each activity by how strongly its own duration swings the project finish date -- the Spearman correlation between the two across all 60,000 runs.
Equipment procurement tops the tornado at 0.50, followed by ramp-up to rate (0.40) and integration & commissioning (0.39). Ramp-up to rate is also on the critical path in 100% of runs. These are the activities to attack first.
Beyond the activity-duration uncertainty, six discrete risk events can each fire and inject extra delay. The Pareto ranks them by expected schedule impact -- probability times mean delay -- and the cumulative line shows how a handful dominate.
The top five of six events carry roughly 80% of the expected discrete-risk delay. A supply-chain component delay leads at 3.0 weeks expected, with equipment downtime at 2.8 weeks and controls integration issues at 2.3 weeks close behind. That ranking is where mitigation budget should go first.
Schedule risk and cost risk are the same risk. Every day beyond the deterministic plan carries overhead, and the discrete risks carry direct costs of their own. The before/after histogram shows the total project cost distribution -- and what targeted mitigation does to it.
Before mitigation the cost averages $33.8M with a P90 of $38.1M, and 25% of runs breach the $36M budget. Mitigation -- predictive maintenance, supplier diversification and workforce training -- pulls the mean to $32.1M, the P90 to $35.6M, and the probability of breaching budget down from 25% to 8%. The gap between the two curves is the value of the risk-management programme, quantified.
Tamara turns an eleven-activity plan that looks tidy on paper into a probabilistic forecast a sponsor can actually commit to. The finish-date histogram replaces "about 535 days" with a centred forecast near 683 days; the Gantt and tornado say where the schedule risk lives; the Pareto ranks the discrete threats; and the cost chart prices the mitigation. Together they convert a single optimistic number into a quantified schedule-and-cost risk position.