Machine learning predictive model for cycle time estimation
An advanced predictive model using machine learning to estimate cycle times, utilizing multivariate feature selection and regression algorithms like random forests. Ideal for manufacturing and industrial applications.
Overview
This solution offers an innovative machine learning predictive model designed to estimate cycle times in industrial and manufacturing settings. By leveraging advanced data science techniques such as multivariate feature selection and sophisticated regression algorithms, this model aims to optimize the prediction accuracy of cycle times. This capability allows businesses to improve operational efficiency and resource allocation.
Technical specifications
- Feature selection: Uses multiple multivariate feature selection methods, such as recursive feature elimination, to identify the most relevant variables that impact cycle time.
- Regression methods: Implements advanced algorithms like random forests and support vector regression to build robust predictive models.
- Data requirements: Requires a high-quality and sufficiently large training dataset for effective model training and validation.
- Customizable: Allows inclusion of client-specified variables, ensuring the model meets specific business needs.
Technology readiness level
The technology is currently at TRL 3, indicating that it has been experimentally validated in a laboratory setting. Future steps include the identification of training and test datasets, feature selection experiments, and model testing to further validate and enhance the predictive capabilities.
About Central Connecticut State University
Central Connecticut State University is a comprehensive public university—the largest in the Connecticut State Colleges & Universities system—located in metropolitan New Britain. It connects with industry through the Central Cooperative Education Program and employer‑aligned internships, as well as noncredit professional education for workforce upskilling. An Institute of Technology & Business Development and on‑campus event services make it easy for companies to access training, facilities, and talent. Innovative Spaces—including the AI Corridor, Cybersecurity Hall, XR Lab, and Nursing Simulation Labs—and business technology labs provide hands‑on environments for collaboration. Proximity to Hartford’s insurance and aerospace corridor further strengthens partnerships. Research is supported through competitive federal and state awards managed by the university’s grants office, and IP is commercialized under a management agreement with UConn.
delivered to your inbox.