Artefact for "Optimal spare management via statistical model checking: A case study in research reactors"

# Artefact for "Optimal spare management via statistical model checking: A case study in research reactors"

This artefact contains the Uppaal model, analysis queries and results for the paper:

*"Optimal spare management via statistical model checking: A case study in research reactors"*
by Reza Soltani, Matthias Volk, Leonardo Diamonte, Milan Lopuhaä-Zwakenberg, and Mariëlle Stoelinga
Published at FMICS 2023 

## Prerequisites
- [Download Uppaal](https://uppaal.org/downloads/):
  We use *Uppaal 5.0* in our evaluation. 
  *Uppaal Stratego 4.1.20* can also be used, but takes significantly more computation time (days instead of minutes).

## Model
Open the provided model file in Uppaal via `File -> Open System` and load the file `Spare_ver4.3.xml`.
The tab `Editor` shows the automata for the individual components (BE, spare, warehouse, etc.) as described in Sect. 5.

## Results
By selecting the tab `Verifier`, the synthesis and analysis queries from Table 3 (in Sect. 6.2) can be seen.
The results from previous runs of these queries are shown and correspond to the results from Table 3.

### Reproduce results
The queries can be rerun by selecting them in order and clicking the button `Check` for each of them.
The first learning query `MinCost` can take several minutes, the other ones finish in seconds.
We also provide queries for spare configurations from 1 to 10 spares.
These queries were used to generate the plots in Figures 9 and 10.

### Configuration
The configuration has already been adapted according to Table 2 and can be seen by clicking the preferences (gear symbol).
The most relevants are the `learning parameters` and `statistical parameters`.

### Changes in the new version 
In the current version (Spare_ver4.3.xml) we use the official release of *Uppaal 5.0*, while we used *Uppaal 5 rc-3 (pre-release)* in the previous version (Spare_ver4.2.xml).
Also, we set the number of the simulation to 100000 for all the *Estimated Value* queries.