Company Specific Proprietary Projects

Company-specific proprietary projects are sponsored by a specific company. The scope and duration are project-specific. The project findings will be protected by an explicit IP agreement negotiated between the university and the company.

Bayesian Predictive Analytics for Part Failures Using Service Data

• Sponsor: Toyota Industries
• Duration: November 2019 – November 2020
• Description:
  In this collaborative project with Toyota, our aim is to understand and utilize Toyota’s service dataset for service event prediction through a series of descriptive and predictive analytics methods. These data analytic methods will help provide valuable information for optimizing service operations at Toyota. The eventual goal is to investigate the feasibility of providing additional analytics tools in order to strengthen their business processes and benefit the bottom line.

Research on Natural Language Processing for Business Process Improvement and Automation

• Sponsor: Jewelry Mutual
• Duration:  September 2019 – August 2020
• Description:In this collaborative project with Jewelry Mutual (JM), we will research on the state-of-art machine learning technique for natural language processing to apply to the business processes at JM. The long-term goal of this collaboration is to establish a set of descriptive, predictive, and prescriptive data analytics methods that are tailored to address both the significant opportunities and the challenging needs of business process improvement. The goal is not only to establish various analytics algorithms, but also to demonstrate the power of data and establish an effective procedure of approaching similar problems for various business processes and platforms.

Analytics for Internet of Things (IoT) Enabled Commercial Kitchen Appliances

• Sponsor: Welbilt
• Duration: June 2018 – June 2019
• Status: Finished
• Description:
  The unprecedented data availability in IoT enabled systems provides significant opportunities for smart data analytics and decision making for operations management but, at the same time, it reveals critical challenges. First, the high dimensional data with heterogeneity and diverse data types often hinders establishing a unified analytics framework. Second, individual-level data has become available in large scale and consequently, there is a pressing need for individualized modeling and analysis. Lastly, the large amount of data provides significant opportunities for realistic data-driven decision making for operations management. Therefore, to truly fulfill the promise of smart and connected systems and facilitate the transformation from data-rich into decision-smart, novel data analytics and data-driven decision-making methods are urgently needed. The long-term goal of this project is to establish a set of data-driven modeling, failure prognosis, and service decision-making methods that are tailored to address both the significant opportunities and the challenging needs of emerging Internet of Things (IoT) enabled systems.

DATA-DRIVEN FAILURE PREDICTIVE ANALYTICS FOR INTERNET OF THINGS (IOT) ENABLED SERVICE SYSTEMS

• Sponsor: Toyota Material Handling North America
• Duration: Nov. 2016 – Oct. 2017
• Status: Finished
• Description:
  It was led by Professors Raj Veeramani, Shiyu Zhou and Kaibo Liu. The overarching goal of the proposal was to establish a series of data-driven modeling, failure prognosis, and service decision-making methodologies that are tailored for both the opportunities and the needs of emerging Internet-of-Things (IoT) enabled service systems.