Document Clustering for ERP Update Compatibility Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for upgrading enterprise resource planning systems are time-consuming and costly due to the need to test large volumes of financial accounting documents for compatibility with new customizing settings, often requiring lengthy simulation phases that can disrupt critical processes and introduce residual risks.
Innovation Solution
A computer-implemented method for document clustering in in-memory databases, where structurally identical documents are grouped and tested in a smaller, representative set, allowing for faster and more reliable simulation of updates without affecting the production system, using attributes like company code, document type, and general ledger accounts to ensure compatibility.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all financial accounting documents are tested through the posting engine with new customizing settings, then compatibility and reliability of the update are improved, but the testing time and implementation duration increase significantly
Solution Approach 1:
The patent segments the large volume of financial accounting documents into smaller groups based on structural characteristics and customizing parameters. By dividing the testing workload into manageable segments, the system can process and validate documents more efficiently while maintaining comprehensive coverage of compatibility scenarios.
Solution Approach 2:
The patent applies partial action by testing a representative subset of documents rather than all documents. The system identifies and tests only the necessary portions of the document set that are sufficient to validate compatibility with new customizing settings, avoiding the excessive time consumption of processing the entire document volume.
2Reliability
If a simulation phase is run for the entire year to identify all incompatible processes, then the reliability of the update is improved, but the productivity and ease of operation are worsened due to extended disruption of critical processes
Solution Approach 1:
The patent performs preliminary action by conducting compatibility testing on a representative sample of documents before full implementation. This preliminary validation provides sufficient confidence to proceed with the update without requiring an extended year-long simulation phase, thereby maintaining reliability while improving implementation speed.
Solution Approach 2:
The patent creates a universal testing approach that uses a representative subset of documents to validate multiple compatibility scenarios simultaneously. By designing test cases that cover various customizing parameters and document types, the system achieves comprehensive validation without the need for extensive prolonged testing.
3Loss of time
If documents are clustered into groups and only one document from each group is tested, then the testing time is reduced, but the measurement precision and manufacturing precision of compatibility testing may be worsened
Solution Approach 1:
The patent applies local quality by selecting specific representative documents from each cluster based on their characteristics and relevance to customizing parameters. Rather than randomly selecting any document from a cluster, the system chooses documents that best represent the local quality and variability within each group, thereby maintaining testing accuracy while reducing overall test volume.
Solution Approach 2:
The patent uses parameter changes by varying the selection criteria for representative documents based on different customizing parameters and document types. This ensures that the selected subset adequately represents the diversity of the full document set across multiple dimensions, maintaining precision while enabling efficient testing.
Data Source
AI summary
A computer-implemented method of testing a compatibility of an update to an enterprise resource planning system that includes clustering documents into groups, where documents are generated from transactional data of an enterprise stored in an in-memory database and have attributes. Groups are stored in memory and include documents that are structurally identical. According to method, one or more processors then add one document from each group into a test set of documents, run a simulation of the update on this test set, and based on the simulation provide data characterizing the compatibility of the update. Related apparatus, systems, techniques and articles are also described.