Comparer des méthodes

	Programmation en nombres entiers bayésienne	Programmation linéaire bayésienne
Domaine	Simulation	Simulation
Famille	Process / pipeline	Process / pipeline
Année d'origine	1990s–2000s	1970s–1980s
Auteur d'origine	Baptiste, Lassagne, Nuijten and others in Bayesian optimization community	Integrated from Dantzig (LP) and Zellner/Bayesian econometrics traditions
Type	Probabilistic combinatorial optimization	Optimization under Bayesian uncertainty
Source fondatrice	Baptiste, P., Lassagne, I., & Nuijten, W. (2001). Bayesian reasoning in mixed integer programming. European Journal of Operational Research, 130(2), 293–313.	Dantzig, G. B. (1963). Linear Programming and Extensions. Princeton University Press, Princeton, NJ. ISBN: 9780691059136
Alias	BIP, Bayesian combinatorial optimization, Bayesian discrete optimization, probabilistic integer programming	BLP, Bayesian LP, Bayesian stochastic linear programming, prior-posterior LP
Apparentées	6	6
Résumé	Bayesian Integer Programming (BIP) integrates Bayesian probabilistic reasoning with integer programming to solve combinatorial optimization problems under uncertainty. Instead of treating parameters as fixed, it encodes prior beliefs about uncertain coefficients and updates them with observed data, producing a posterior-guided search over integer-feasible solutions. The approach is widely used in scheduling, resource allocation, and supply-chain planning where data are incomplete or noisy.	Bayesian Linear Programming (BLP) integrates Bayesian statistical inference with classical linear programming to handle uncertainty in model parameters such as objective function coefficients, constraint coefficients, or right-hand-side values. Instead of treating parameters as fixed or governed by worst-case bounds, BLP uses prior beliefs updated by data to form posterior distributions, which then guide the LP formulation and solution, producing decisions that are optimal in a probabilistic, data-informed sense.