We published an article that makes the case that rules are excellent for expressing business logic, constraints are excellent for defining and searching the space of feasible solutions, and integrating both within a single decision model can make optimization simultaneously powerful and accessible to business users.

The article describes the evolution of this idea in OpenRules. An earlier Rule Solver transformed rule-based models into constraint-satisfaction problems, but practical limitations often forced rule and constraint portions into separate decision services. The redesigned 2026 OpenRules Rule Solver introduces an expression language that allows known and constrained/unknown variables, ordinary rules, global constraints, and constrained expressions to coexist directly in traditional decision tables. Link