Procurement Price Anomaly Detection for Alternate Buying Options
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional procurement systems lack a comprehensive view, domain-specific insights, and predictive capabilities to identify and prevent anomalous transactions, leading to sub-optimal pricing and significant revenue loss due to procurement frauds and inefficiencies.
Innovation Solution
A method and system that utilize influencer graphs, clustering algorithms, and decision trees to analyze procurement data, identify influential features, detect price anomalies, and recommend alternate buying options based on seasonal behaviors and maverick transactions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional statistical tools and BI tools are used for procurement analysis, then analytics reports can be generated, but the analysis is done in silos lacking comprehensive view and domain-specific insights
Solution Approach 1:
The patent combines multiple analytical components (influencer graph analysis, seasonal behavior detection, price anomaly identification, clustering algorithms) into a unified procurement analysis system. This integration merges previously siloed analyses into a comprehensive view that captures interrelationships between different procurement factors, enabling holistic insights while managing complexity through systematic architecture.
Solution Approach 2:
The system performs multiple functions within a single analytical framework: it identifies influential features, detects seasonal patterns, finds price anomalies, clusters transactions, and generates recommendations. This multi-functional approach eliminates the need for separate analytical tools for each function, providing comprehensive procurement insights through a unified system that handles diverse analytical tasks.
2Ease of operation
If conventional methods are used for procurement analysis, then observations can be provided, but actionable recommendations are not generated
Solution Approach 1:
The system closes the loop between analysis and action by generating specific, actionable recommendations based on detected anomalies and patterns. The feedback mechanism translates analytical findings into practical procurement guidance, such as identifying alternative suppliers or optimizing purchase timing, enabling users to directly implement improvements based on system insights.
Solution Approach 2:
The system performs preliminary analysis to identify potential issues and opportunities before procurement decisions are finalized. By detecting price anomalies, seasonal patterns, and influential factors in advance, the system enables proactive procurement optimization rather than reactive problem-solving, allowing organizations to take preventive actions that save costs.
3Reliability
If conventional reactive methods are used, then current issues can be identified, but predictive capabilities are lacking
Solution Approach 1:
The system performs preliminary analysis to identify potential issues and opportunities before procurement decisions are finalized. By detecting price anomalies, seasonal patterns, and influential factors in advance, the system enables proactive procurement optimization rather than reactive problem-solving, allowing organizations to take preventive actions that save costs.
Solution Approach 2:
The system incorporates dynamic elements by detecting seasonal behaviors and time-varying patterns in procurement data. The analytical model adapts to changing conditions by identifying temporal patterns and adjusting predictions based on seasonal fluctuations, enabling the system to provide accurate predictions despite varying market conditions and procurement patterns.
4Quantity of substance
If large numbers of transactions are analyzed, then comprehensive coverage is achieved, but difficulty in identifying procurement frauds increases
Solution Approach 1:
The system extracts and focuses on the most critical features and patterns from large volumes of procurement transactions. By identifying influential features through graph analysis and extracting key anomalies using clustering algorithms, the system filters out noise and focuses computational resources on the most significant deviations, making fraud detection feasible even at scale.
Solution Approach 2:
The system replaces manual review processes with automated analytical methods including influencer graphs, seasonal detection algorithms, and clustering techniques. This substitution enables the systematic analysis of large transaction volumes that would be impossible to review manually, using computational algorithms to identify patterns and anomalies across extensive procurement data sets.
Data Source
Figure 1A
Figure 1B
Figure 2A
AI summary
A general notion in the procure-to-pay domain is that unit price of an item is dependent on the quantity being purchased. Conventional methods fails to provide predictive model considering a comprehensive view of the core problems and domain specific insights. Initially, the system receives procurement data and identifies influence values of product features. Further degree of impact of features are identified and highly influential features are identified based on that. Further, seasonal behaviors are identified, and price anomalies are identified based on that. Further, an optimal cluster is identified based on a silhouette score associated with each of the plurality of candidate clustering algorithms. Further, anomalous transaction clusters are identified. Further, a plurality of reasons associated with the identified anomalous transaction clusters are identified based on maverick transactions using decision trees and an alternate buying option is recommended for a new procurement order based on the identified plurality of reasons.