Completeness Graph Workflow Automation Without External Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing workflow automation and data orchestration technologies face challenges such as sequential execution plans, external synchronization requirements, complex error handling, and unscalable model definitions, leading to inefficiencies and inconsistencies.
Innovation Solution
A declarative completeness graph integrated with the managed systems, using a decision-making algorithm to determine attribute values and modify the entity state, allowing for efficient and scalable workflow automation and data orchestration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If workflows are modelled as sequential steps in an execution plan, then the workflow execution is structured and controllable, but the workflow definition becomes complex and unscalable
Solution Approach 1:
The patent segments the workflow model into two independent parts: (1) a declarative business model defining entities, attributes, and relationships, and (2) an execution engine that automatically generates workflow steps based on the model. This segmentation eliminates the need for complex sequential execution plans while maintaining structured control, as the execution engine derives the workflow structure automatically from the declarative model.
Solution Approach 2:
Instead of defining workflows as sequential steps and deriving the model from them, the patent inverts the approach by defining the declarative business model first and having the execution engine automatically generate the workflow steps. This inversion simplifies workflow definition while maintaining execution control, as the model becomes the single source of truth.
2Extent of automation
If workflows are defined externally to the managed system, then workflow automation is achieved, but continuous synchronization between the workflow engine and the system is required
Solution Approach 1:
The patent merges the workflow definition with the business model by embedding the declarative model directly within the managed system. The execution engine accesses the model locally without requiring external synchronization, eliminating time losses associated with keeping external workflow definitions in sync with system changes.
Solution Approach 2:
The declarative business model acts as an intermediary between the execution engine and the managed system. It provides a local, self-contained representation of business logic that the execution engine can process without continuous external synchronization, while still reflecting the current state of the managed system.
3Reliability
If the workflow engine maintains the state of each running workflow instance, then workflow execution tracking is achieved, but data duplication and increased possibility of inconsistencies occur
Solution Approach 1:
The patent extracts workflow state management from the execution engine and stores it directly in the managed system's database. This eliminates data duplication between the workflow engine and the system, as both now access the same centralized state information, thereby maintaining consistency while preserving execution tracking reliability.
4Reliability
If error handling rules and fallback options are explicitly defined in the execution plan, then error cases are handled, but the workflow definition process becomes complex and unscalable
Solution Approach 1:
The patent incorporates error handling rules and fallback options directly into the declarative business model definitions, rather than adding them separately to execution plans. This preliminary integration of error handling logic into the model itself simplifies workflow definitions while maintaining comprehensive error handling capabilities, as the execution engine automatically applies these pre-defined rules.
Data Source
AI summary
A method may include obtaining a completeness graph including conditions applied to attributes of an entity, determining that a missing attribute subset of the attributes lacks a corresponding value, determining that a condition corresponding to a missing attribute of the missing attribute subset is satisfied, and obtaining a value for the missing attribute. The value may be a result of performing a task in the workflow. The method may further include modifying a state of the entity by assigning the value to the missing attribute to obtain a modified state of the entity.