Manufacturing with DAPI
Manufacturing processes can be improved through data analytics by identifying relevant variables within unique processes with a goal of reducing the time needed to create products and reduce waste.
- Professor Myong K. Jeong has received research funding from the National Science Foundation, IEEE, Korea Institute of Science and Technology, Electronics and Telecommunications Research Institute, Samsung Global Research Program, and other organizations to create more efficient mass-production systems. Patent Citation Analysis for Technology Management , Manufacturing Sensor data Analytics: Spatial Data Modeling
- Professor Susan Albin is preparing a study to identify the most relevant variables in a manufacturing process. At present there are numerous extraneous variables that are not relevant to streamlining these processes, but they are difficult to distinguish from the ones that are used. Through the use of test environments which limit the number of variables present, the team will work to determine which ones are important to the process. Data Mining: Feature Selection for Batch Production
- Glass product production can be greatly improved by reducing variabilities in the manufacturing process through methodologies to model, monitor, and control the process. Professor Susan Albin is working with Corning Inc. to develop advanced display glass suitable as a substrate for electronics as small as smartphones to as large as television sets. Sensors are being used for online analysis to determine the most efficient ways to uniformly create the glass product and reduce the number of issues that can disrupt manufacturing. Multivariate Statistical Process Control & Data Mining in Display Glass Manufacturing
- Defect monitoring of high-power, high-capacity batteries during manufacturing reduces the number of batteries requiring manual inspection. Professor Weihong Guo is developing efficient systems to monitor high-power, high-capacity batteries during their manufacturing process to ensure there are no defects in the final products. At present these batteries are used in electric vehicles, and need reliable connections between different batteries. The system being developed will analyze the signal produced during the ultrasonic welding process in the battery's production, noting variances between the welding time and the final product to see what the optimal welding time is. Initial results from a case study have led to an 80 percent defect reduction rate.
- Professor Weihong Guo is developing new sensor fusion methods to help improve online process monitoring and fault diagnosis. These new methods are able to handle heterogeneous sensor data efficiently to facilitate real-time decision-making in manufacturing processes.
Improving Healthcare with DAPI
Utilizing the power of data analytics in healthcare allows for more efficiency related to patient care and well-being. Advanced smart systems will allow for learning from previous successes and failures, developing consistencies in determining the best outcome for the patient every time.
- Professor Susan Albin is working with professor Kang Li and the Robert Wood Johnson Medical School to determine more efficient ways to prevent patients from requiring hospitalization. The research analyzes how population care coordinators can reduce re-hospitalizations by monitoring their patients remotely, utilizing phone calls and other technologies to ensure these patients are treated in a timely fashion after their initial hospitalization. Health Service Systems: Role of Population Care Coordinators in Family Practice, Classification of Uncertain Data Using Group Probabilistic Distance Measure
- Professor Weihong Guo is developing a system to predict whether a patient will need surgery for a rotator cuff tear. These surgeries often result in long periods of physical therapy, where patients are in pain and require further treatment. The system will help doctors determine whether a surgery is even necessary by analyzing patient information earlier than is done at present.
Intelligentizing Transportation Systems and Infrastructure Design with DAPI
Intelligent traffic and transportation systems on the nation's largest roads will contribute to vital advances in infrastructure design and technology. At present these systems are inefficient and difficult to implement. Improved systems using sensors in specific locations and developed algorithms will contribute to benefits in congestion relief, increased travel efficiency, automotive advances, and safety.
- Professor Myong Jeong is creating smart transportation monitoring systems, which can utilize the internet to locate collisions and let vehicles predict traffic slowdowns due to construction or other issues. Automatic incident detection systems on highways can also alert emergency services when collisions do occur, ensuring their timely arrival and reducing the length of time that the roads are blocked.
Advancing Technology with DAPI
State-of-the-art sensors are used for data collection and allow for system improvements. Each iteration of improvements brings the process closer to its maximum efficiency, saving time, energy and resources.
- Professor Myong Jeong is using "big sensors" to monitor semiconductor, cybernetic and patent citation manufacturing and design processes to locate inefficiencies in these systems. The data will be used to eliminate the issues, resulting in resources being allocated better to maximize the output. The data can also be used to more quickly locate damaged semiconductor wafers, ensuring they are easier to find and harder to mix with a shipment of products.