Multi-Target ML Model for Product Execution Outcome and Lifespan Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing project management tools lack the intelligence to accurately predict the outcomes and lifespan of products/projects at the planning stage, especially with the rapid evolution of technology and business factors.
Innovation Solution
A multi-target machine learning (ML) model, such as a multi-output deep neural network (DNN), is trained using historical product execution and lifespan data to predict both the execution outcome and lifespan estimate of new products, enabling better-informed decision-making.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional project management tools are used, then ease of operation is maintained, but prediction accuracy of product execution outcome and lifespan is insufficient
Solution Approach 1:
A multi-target machine learning model serves as an intermediary between historical product data and future execution outcomes. The model processes multiple input features (product type, business domain, language, database, consumption, deployment) and simultaneously predicts both execution outcome and lifespan, resolving the contradiction by introducing an intelligent mediation layer that enhances prediction accuracy without requiring direct complex manual analysis
Solution Approach 2:
The machine learning model performs multiple prediction functions simultaneously - it predicts both execution outcome (classification) and lifespan (regression) from the same input features. This multi-functionality allows a single system to address multiple prediction needs, improving overall measurement precision while avoiding the need for separate complex analysis tools for each prediction type
2Productivity
If multiple prediction targets are addressed simultaneously, then productivity is improved through comprehensive insights, but device complexity increases due to multi-target modeling
Solution Approach 1:
The patent combines multiple prediction tasks (execution outcome prediction and lifespan prediction) into a single multi-target machine learning model. By merging these functions into one unified system that processes shared input features simultaneously, the solution improves productivity through comprehensive insights while managing complexity through shared computational infrastructure and feature processing pipelines
3Measurement precision
If historical data from multiple dimensions is utilized, then measurement precision of predictions is improved, but loss of time in data processing increases
Solution Approach 1:
The system performs preliminary processing of historical product data during the training phase, pre-computing feature representations and relationships across multiple dimensions (product type, business domain, language, database, consumption, deployment). This preliminary action stores learned patterns and correlations in the trained model, enabling fast predictions on new products without re-processing all historical data, thus improving measurement precision while minimizing time loss during actual prediction operations
Data Source
AI summary
An example methodology includes, by a computing device, receiving information regarding a product from another computing device and determining one or more relevant features from the information regarding the product, the one or more relevant features influencing predictions of a product execution outcome and a lifespan estimate. The method also includes, by the computing device, generating, using a multi-target machine learning (ML) model, a first prediction of an execution outcome of the product and a second prediction of a lifespan estimate of the product based on the determined one or more relevant features, and sending the first and second predictions to the another computing device.