Table 6. Statistical analysis of the model checking results for the rat cardiovascular system dynamics, the uterine contractions of labour, the Xenopus laevis cell cycle, and the acute inflammation of the gut and lung case studies.
| MId | SId | % true PBLMSTL | #total MSTML | #true MSTML | #false MSTML | Execution time | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| μ | σ | μ | σ | μ | σ | μ | σ | |||
| 1 | 1 | 100 | 1 | 0 | 1 | 0 | 0 | 0 | 17.67 | 0.12 |
| 2 | 100 | 1 | 0 | 1 | 0 | 0 | 0 | 17.61 | 0.13 | |
| 3 | 100 | 1 | 0 | 0 | 0 | 1 | 0 | 17.8 | 0.36 | |
| 2 | 4 | 100 | 1 | 0 | 1 | 0 | 0 | 0 | 0.55 | 0.01 |
| 5 | 100 | 1 | 0 | 0 | 0 | 1 | 0 | 0.54 | 0.01 | |
| 6 | 100 | 1 | 0 | 1 | 0 | 0 | 0 | 0.54 | 0.01 | |
| 3 | 7 | 100 | 28.79 | 2.04 | 28.61 | 1.62 | 0.19 | 0.44 | 35.35 | 2.44 |
| 8 | 100 | 28 | 0 | 28 | 0 | 0 | 0 | 34.29 | 0.09 | |
| 9 | 100 | 28 | 0 | 28 | 0 | 0 | 0 | 35.36 | 0.99 | |
| 4 | 10 | 100 | 28 | 0 | 28 | 0 | 0 | 0 | 87.39 | 0.72 |
| 11 | 100 | 28 | 0 | 28 | 0 | 0 | 0 | 90.27 | 2.23 | |
| 12 | 100 | 28 | 0 | 28 | 0 | 0 | 0 | 87.03 | 0.65 | |
Entries in the “MId” and “SId” columns represent the numeric identifiers associated with each computational model and its corresponding PBLMSTL statements. The “% true PBLMSTL” column describes what percentage of the 500 model checker executions concluded that the PBLMSTL statement is true. “#total MSTML” represents the total number of MSTML files evaluated for the PBLMSTL statement during a single model checker execution; columns “#true MSTML” and “#false MSTML” represent the number of MSTML files for which the PBLMSTL statement was evaluated true, respectively false, during a single model checker execution. “Execution time” records the average runtime in seconds for each model checker execution. “μ” and “σ” represent the mean and standard deviation. Due to the deterministic nature of computational models M1 and M2 only one simulation trace was employed for their verification (see table rows corresponding to MId 1 and MId 2, table column 4). Conversely the number of simulation traces considered for the verification of computational models M3 and M4 was equal to ≈28 (see table rows corresponding to MId 3 and MId 4, table column 4), and was computed as a function of the input parameters α and β of the improved statistical hypothesis testing model checking approach. The model simulation traces employed for the verification of computational models M3 and M4 were chosen randomly from the collection of 1500, respectively 500 simulation traces generated to compute the average execution times given in Table 4.