Workflow Completion Time Prediction Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Enterprises face challenges in providing a clear timeline for account opening due to varied reasons leading to delays, especially in product origination processes involving multiple agent personas across different departments, resulting in a lack of transparency and real-time status during product origination.
Innovation Solution
Utilizing machine-learning models to predict account opening dates by considering user data, workflow stages, and handling real-time data fluidity, enabling accurate and transparent predictions of completion times based on dynamic workflow positions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning models are used to predict completion times, then prediction accuracy is improved, but system complexity increases
Solution Approach 1:
The patent introduces a machine learning model as an intermediary component that sits between the workflow management system and the customer service interface. This model processes workflow data and generates completion time predictions, thereby improving prediction accuracy while isolating the complexity within a dedicated predictive analytics module rather than分散 across the entire system.
Solution Approach 2:
The system is segmented into distinct functional modules: data collection from workflow stages, feature engineering, machine learning model processing, and prediction output. This segmentation allows each component to be optimized independently and simplifies system maintenance while maintaining high prediction accuracy through specialized processing at each stage.
2Loss of information
If real-time data from multiple workflow stages is collected, then prediction transparency is improved, but data processing complexity increases
Solution Approach 1:
The patent extracts only the essential features from the complex workflow data that are most predictive of completion times. Rather than processing all available data from multiple stages, the system identifies and extracts key indicators such as current stage position, historical duration patterns, and bottleneck metrics, thereby maintaining prediction transparency while reducing processing complexity.
Solution Approach 2:
The system transforms raw workflow data into standardized parameters and features suitable for machine learning processing. This includes normalizing time durations, encoding stage positions, and aggregating metrics across different workflow stages, which simplifies the data structure while preserving the information needed for transparent and accurate predictions.
Data Source
AI summary
Machine learning techniques are disclosed for predicting a task event such as a service completion event based on a predefined workflow. In one aspect a method includes obtaining initial data for a service request (e.g., an account application), enriching the initial data with data from one or more repositories of an enterprise executing the service request, generating a data structure comprising independent variables extracted from the enriched data, receiving a request for a prediction of a completion time for the service request (e.g., an account opening event) at a first time during processing of the service request in accordance with each workflow, in response to receiving the request for the prediction, inputting the data structure into a machine-learning regression model, predicting, using the machine-learning regression model, a completion time for the service request, and providing the completion time for the service request.