Cost Allocation Estimation Using Machine Learning and Sender-Receiver Matrices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large enterprises face prolonged processing times for cost allocations, often exceeding 24 hours, which can be inefficient and resource-intensive, especially when accurate results are not needed in all cases.
Innovation Solution
Implementing a computer-implemented method using machine learning to estimate cost allocations by consolidating transactions into sender-receiver totals, determining sender-receiver percentages, and calculating estimated allocations based on current actual costs and historical data, thereby reducing processing time and maintaining acceptable accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional cost allocation processing is used to ensure accurate results, then measurement precision is improved, but loss of time increases significantly
Solution Approach 1:
The patent applies partial action by using machine learning to estimate cost allocations for cases where full accuracy is not required, processing only the necessary portion of transactions through traditional methods while using ML estimations for others, thereby reducing overall processing time while maintaining acceptable accuracy levels
2Reliability
If traditional cost allocation processing is used to maintain reliability, then reliability is improved, but productivity deteriorates
Solution Approach 1:
The patent introduces machine learning models as an intermediary between traditional cost allocation processing and final results. The ML models pre-process and estimate cost allocations, serving as a mediator that reduces the computational burden on traditional processing systems while maintaining reliable outcomes through validation against actual transactions
3Measurement precision
If comprehensive transaction processing is performed to improve measurement precision, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the cost allocation processing system into distinct components: a machine learning estimation module that handles preliminary processing, a traditional processing module for validation and adjustment, and a reconciliation module. This segmentation allows each component to specialize in specific tasks, improving overall precision while managing system complexity through modular architecture
Data Source
AI summary
The disclosure generally describes methods, software, and systems for estimating cost allocations, including a method for the following steps. Using a machine learning system, transactions are consolidated into estimated sender-receiver totals for costs transmitted by senders to receivers in an organization. A sender-receiver percentage matrix is determined from the estimated sender-receiver totals of a given sender and for each receiver of the transactions from the given sender. The sender-receiver percentage matrix includes, for each sender, estimated sender-receiver percentages. Current actual costs are determined for each sender to receivers for a given time period. Estimated cost allocations are determined for given time period using the sender-receiver percentage matrix. The estimated cost allocations are determined for each receiver in the organization based on a function of the current actual costs for each sender. A report that includes the estimated cost allocations is provided for presentation to a user.