Welcome!
FelooPy is a Python package for user-friendly decision-making across complex systems, industries, and supply chains. It provides an easy, unified, and vendor- and provider-neutral API to access numerous other APIs and powerful open-source (free/non-proprietary) or closed-source (proprietary demos/full versions) solvers and algorithms to obtain optimal, near-optimal, or best-possible decisions across three ORbits:
- ORbit 1: Uncertainty modeling: deterministic (D), stochastic (S/TS/MS), fuzzy (F), robust (R/RO/DRO/DDRO), and their combinations.
- ORbit 2: Objective and criteria support: single-level/objective (SL/SO), bi-level/objective (BL/BO), multi-level/objective (ML/MO), many-level/objective (ML/MO), and their combinations.
- ORbit 3: Search-space exploration and exploitation: linear (LP), nonlinear (NLP), integer (IP), mixed-integer linear (MIP/MILP), mixed-integer nonlinear (MIP/MINLP), constrained (CP), alternative (MADM/MCDM), landscape (X-Heuristics), and state-space (DRL/SDM) search, and their combinations.
With FelooPy, formulating, coding, solving, analyzing, and benchmarking mathematical, computational, or statistical models is streamlined. It promotes shareability and reproducibility among users with varying access levels to free or proprietary solvers and algorithms. Models are passed to environments that are searched iteratively, sequentially, or in parallel—each integrated to determine how operations should be performed in real or virtual systems, industries, or supply chains. Its modular environments allow experts to reuse capabilities instead of starting from scratch for each use case.
Visit our INFORMS OR/MS Tomorrow article for more information.
For online discussions of feloopy or to ask a question, you can use the comment composer below this screen!
If you use FelooPy in your work, please cite as:
Or if you prefer to cite its GitHub repository: