Software Reliability Growth Model (2D)

Every test campaign finds bugs fastest at first, then slower and slower as the obvious defects run out — a pattern software reliability engineers model as a non-homogeneous Poisson process. This 2D chart simulator seeds a testing run by drawing a random total defect count from a Poisson distribution and giving each defect an independent, exponentially-distributed discovery time, then plays the test clock forward: a growing bar chart shows the actual, stochastic cumulative defect count against the smooth Goel-Okumoto mean-value curve it fluctuates around, with every individual discovery marked on its own timestamp. Drag to pan and scroll to zoom into any stretch of the timeline, adjust the ultimate defect count and detection rate to see how codebase size and test-suite maturity reshape the discovery curve, and watch the live discovery-rate and remaining-defect readouts track a real quality-assurance dashboard.