* Remove unused parameters
A previous commit disconnected bonsai-related parameters from the Connector but mistakenly did not delete them from the model (they appear in the experiment properties, as if still functional)
* Improve comments & code
- Remove unused code
- Add small workaround code for locally hosted sims
- Add/edit clarifying comments
* Remove stop options from sim
Stop/terminal conditions offset entirely to Inkling, as current AnyLogic integration does not support this (may cause training issues for users who are experimenting)
* Clean/fix inkling
- Improve comments: removed an inaccuracy (reference to some nonexistent "ModelAction" class), consistency of where they're written, clarified purposes
- Replace 'Action' type with 'SimAction' type (the latter was previously unused)
- Remove unused "Math" import
* Update README.md
- Removed a sentence only applicable to the original simulation model
- Replaced explanations about the old, "wrapper" workflow with the new RL experiment workflow
- Updated explanations about how to perform local assessments
- Updated sample inkling based on previous commit
- Added more detailed instructions
- Clarified various parts
- Cleaned some grammar
* Update exported.zip
Exported with the latest changes
* Minor comment updates
* Update README.md
- Manually merge some phrases made by @mzat-msft, adjusting slightly for grammar
- Reapply line width
* Rename file/model
* Remove debugging experiments
These were used in the model-exploration stage of bonsai integration to find the ranges of added elements (namely the "rolling" stats) and to gauge how well a static optimization would perform.
None are relevant to the distributable model.
* Remove unused objects
These objects were used during initial testing for brain integration, but not incorporated in the end.
This includes parameters for initial MC state (reverted to their default values) and statistical elements (cumulative data tracking; replaced with rolling data).
* Replace cost parameters with constant variables
Cost values were only intended to be configurable when first adding the 'cost' component to the model.
* Update experiment screens
- Update RL-related parameters to hide themselves in the experiments when not relevant
- Remove unused control elements in OriginalSimulation
- Add relevant controls in OriginalSimulation and PlaybackSimulation
- Add explanatory text in each experiment
- Fix model description in OriginalSimulation
- Fix titles in PlaybackSimulation and TrainingSimulation
* Clean up model
- Rename some variables
- Remove unused code
- Rearrange objects
- Resize some shapes
* Fix bug in initialization
Due to removed parameters in earlier commit, MCs were being initialized to their default values for vehicles/prod. rate (both 0), causing inkling errors.
* Remove now-unused configuration variables
* Simplify inkling reward functions
- Made the reward functions for "turnaround" and "utilization" generic, consolidating a total of 7 functions into 2.
- Reorganized some function logic for clarity and conciseness.
* Remove inkling's time-based constraints in observation
- Constraints were assuming model would always run for ~30 days. However, since constraints still are enforced in exported brains, this artificially limited how long models could be run in playback.
- Time was removed from observable, as it doesn't factor into the logic
Testing showed removing these had a positive impact (comparing final mean scores: 4.23 vs 6.70)
* Correct types in inkling
Some type declarations were different for the same element; chosen types were set based on how they're used in the model - "open" = nominal (boolean), "trucks" = discrete (int), "rate" = continuous (float).
Testing showed this had a negligible effect on brain performance.
* Rename variables in sim/inkling
Trucks -> Vehicles
- Both were used interchangeably; kept with the original model's usage of "vehicles"
Open -> Accepting
- More accurate terminology (e.g., non-accepting MCs still process orders internally, which "non-open" would imply otherwise)
* Condense RL spaces using arrays
This makes code simpler and more compact.
Also renamed some variables to better indicate what it reflects and to include time units.
* Add comments to inkling
Mostly used previous comments; added some new ones
* Minor fixes + export model
- Grammar fix in inkling
- Fix sliders in sim
* ABCA updates, including writeup
* update exported model
* adding docker support for SAs
* adding docker for SAs
* dockerfile
* Delete dockerfile
Co-authored-by: unknown <davidhcoe@msn.com>
Tyler
Marco Zatta
David Coe
davidhcoe
unknown
Evan Kramer
Howard Nager
David Coe