A new-drug programme is one of the longest, most uncertain schedules in any industry - a decade or more from preclinical to NDA filing, with patient recruitment and regulatory phases that routinely run long. A deterministic Gantt collapses all that uncertainty into a single submission date; the reality is a broad distribution. This case study runs a Tamara schedule-risk model of an 11-activity oncology programme to draw the submission-date distribution and read the exposure off its tail.
The submission-date distribution is the headline. The deterministic plan lands at 3,350 days (9.2 years), but the simulated P50 is 3,936 days (10.8 years), the P80 is 4,207 days (11.5 years) and the P90 is 4,353 days (11.9 years). The P10-P90 spread is 810 days - more than two years of uncertainty - and the deterministic plan is beaten in only 2% of runs.
Eleven activities with three-point PERT durations on a critical-path network, spanning preclinical, the three trial phases, CMC and the NDA filing, plus six discrete risk events - recruitment shortfalls, protocol amendments, safety holds, CMC scale-up, site activation and DSMB pauses. One Monte-Carlo of 60,000 iterations drives every chart.
The histogram makes the headline unmistakable: the most-likely outcome is roughly a year past the deterministic plan, and the right tail adds another year on top. Reaching P80 requires 857 days of contingency beyond the plan.
The tornado ranks phases by how strongly their duration correlates with the submission date. Phase III patient recruitment leads at a cruciality of 0.58, ahead of Phase II proof-of-concept (0.41) and Phase I dose escalation (0.39) - certainty in recruitment moves the filing date more than any other lever.
Cost is schedule-linked through burn rate: every day past plan burns about $0.12M of trial infrastructure and team overhead. Before mitigation the cost mean is $615M with a P90 of $697M against a $632M budget - a 39% overrun chance. The mitigation case (specialist recruitment CRO, adaptive design, early parallel CMC) pulls the mean to $588M and the P90 to $661M, cutting the overrun probability to 21% and the P80 finish from 4,207 to 3,974 days.
The contingency ladder reads as a decision table for the portfolio commitment: P50 needs +586 days beyond plan, P70 +754, P80 +857, P90 +1,003 and P95 +1,125 days. Even the median outcome carries nearly two years of buffer over the deterministic 3,350-day plan.
The Pareto ranks the discrete risk events by expected schedule impact. Patient recruitment shortfall dominates at 5.1 expected months - more than double the next event, CMC scale-up failure at 2.4 months; the top five of six events carry ~80% of the expected discrete-risk delay.