Regelcomputers

Regelcomputers, or rule computers, are computing systems that implement explicit, codified rules to guide data processing and decision making. They operate by maintaining a rule base of if-condition–then-action rules and using an inference engine to apply those rules to known facts in working memory, producing new facts or triggering actions.

Originating in the field of artificial intelligence during the 1970s and 1980s, rule-based systems such as OPS5 and CLIPS popularized the approach. In European literature the term regelcomputer emphasizes the rule-driven nature of these systems. The paradigm is commonly described as forward chaining: starting from known data, the engine derives consequences by repeatedly applying applicable rules.

Typical architecture includes a working memory for facts, a rule base, an inference engine, and a rule execution mechanism with a conflict resolution strategy to decide which rule to fire when several are eligible. Pattern matching, often implemented via the Rete algorithm, is central to efficient rule evaluation.

Applications include expert systems, decision-support, business rules management, and regulatory compliance, where transparent, auditable reasoning is valuable. Advantages include explicit control over logical inferences and ease of updating rules without changing program code. Limitations include maintenance challenges for large rule bases, potential performance bottlenecks, and difficulties in handling uncertain or probabilistic knowledge.

Today regelcomputers influence modern rule engines and business rules management systems, and continue to be used in domains requiring deterministic logic and traceability, even as many AI systems blend rules with statistical learning.