Goal-Driven Rule Execution Engine for Business Logic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing rule-based systems face challenges in managing large, complex sets of data-driven rules, which are difficult to understand and maintain, and often require immediate adaptation to data changes, leading to inefficiencies and unpredictability in execution.
Innovation Solution
Transforming data-driven rules into goal-driven rules that can be executed eagerly using a forward-chaining rule engine, preserving the data-driven behavior and allowing for the encapsulation of high-level business concepts like customer creditworthiness, while enabling execution in a unified manner that simplifies logic and improves processing speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If data-driven rules are used to enable immediate adaptation to data changes, then responsiveness is improved, but rule set complexity and difficulty to maintain increases
Solution Approach 1:
The patent segments the rule execution process into two distinct phases: a compilation phase where data-driven rules are transformed into goal-driven rules with explicit dependencies, and an execution phase where the transformed rules are processed. This segmentation separates the complexity of rule transformation from rule execution, making the system more manageable and maintainable while preserving responsiveness.
Solution Approach 2:
The patent applies preliminary action by performing rule transformation and dependency analysis during a compilation phase before actual rule execution. The data-driven rules are pre-processed to identify all data dependencies and transform them into goal-driven rules with explicit dependency clauses, so that during execution, the system can efficiently evaluate rules without performing complex analysis in real-time.
2Speed
If data-driven rules are executed eagerly to provide immediate computation, then processing speed is improved, but the Rete bottleneck reduces overall efficiency
Solution Approach 1:
The patent inverts the traditional approach by transforming data-driven rules (which are executed eagerly) into goal-driven rules with explicit dependencies. Instead of eagerly evaluating all data-driven rules and hoping to find relevant ones, the system lazily evaluates only the specific goal-driven rules that are needed, based on explicit dependency relationships. This inversion eliminates the Rete bottleneck while maintaining the ability to provide immediate computation when needed.
3Ease of operation
If goal-driven rules are used to encapsulate high-level concepts, then comprehensibility is improved, but execution flexibility is reduced
Solution Approach 1:
The patent creates a universal rule transformation framework that can handle both data-driven rules and goal-driven rules within a single system. The transformation process is general and applicable to any rule set, allowing the system to maintain comprehensibility through goal-driven rule encapsulation while preserving execution flexibility through the configurable dependency evaluation mechanism. The same framework adapts to different rule types and execution scenarios.
Data Source
AI summary
Data-driven rules are eliminated from an application, and replaced with goal-driven rules that formally and completely characterize high-level business concepts, while enabling the resulting goal-driven rules to be executed eagerly using a forward chaining rule engine in a way that preserves the data-driven behavior of the original rules. A computer-implemented method for executing business rules provides a unified rule that encapsulates a goal-oriented rule with triggering information selected from dependencies of the goal-oriented rule. The unified rule may be eagerly executed in response to changes in data specified by the dependencies, or lazily executed in response to a request for the value computed by the rule.