Session: DTM-02-01: Artificial Intelligence in Design

Paper Number: 116912

116912 - Use of Data-Mined Customer Reviews to Inform Design Evaluation and Testing of Bassinets 

This paper describes a method to use data-mined customer reviews to identify potential product defects.  The process involves locating negative reviews of a specific product and then extracting comments that have potential connections with the product design.  The extracted comments are then categorized and correlated with features of the product.  Given that the customer comments are generally not specifically tied to engineering requirements, or even posed in engineering terms, this correlation requires some degree of engineering analysis to establish a correlation.  After these complaint-feature correlations are established, then engineering tests directed at understanding the possible defects are performed.  This data-mining process effectively harnesses the massive amount of in-situ testing and evaluation that is performed by customers of the product.  The process is illustrated via a case study of an infant bassinet.  Customer reviews of the product were studied to identify and categorize key complaints about the bassinet.  These complaints were correlated with features and potential defects in the product, including assembly difficulties as well as a sleeping surface that tilts and causes infants to roll and press into the mesh side wall while sleeping.  Then, assembly and sleeping surface deflection tests were conducted.  The sleeping surface deflection tests presented include investigations of how bassinet leg separation affects the deflection and measuring the deflection over time.  The results of the engineering testing confirm the presence of defects in the bassinet and its assembly instructions as suggested by the reviews, and this case study illustrates how data-mined customer reviews provide a valuable source of engineering data and indications of product defects.

Presenting Author: William Singhose Georgia Institute of Technology

Presenting Author Biography: After completing a post-doctoral appointment at MIT, William E. Singhose joined the School of Mechanical Engineering faculty at the Georgia Institute of Technology in 1998. Singhose's research focuses on the dynamics and control of flexible machines. The performance of many types of machines is limited by the vibration and deflection that occur during their operation. For example, when a crane is used to transport material, the payload often oscillates when it reaches the desired location. The workers must then wrestle with the payload to bring it to rest, which slows down the transfer process and also presents a safety hazard. Singhose has developed a control scheme that alters the voltages sent to the motors of cranes, so that the human operator is able to move the payload without oscillation. In a more general sense, his work attempts to predict and minimize the damaging effects of machine flexibility. Simple models are used to capture the important dynamic effects. These models are then used to develop special commands for moving the machine about its workspace without vibration. Using this approach, he has also implemented control schemes on silicon handling robots, scanning devices, and many types of automated manufacturing equipment. In addition to command generation, Singhose's research projects include design of vibration absorbers, spacecraft thruster control, robot manipulator design and control, minimization of friction effects, and compensation for actuator saturation.

Authors:

William Singhose Georgia Institute of Technology
Christopher Adams Georgia Institute of Technology
Anjnee Rana Georgia Institute of Technology
Dooroo Kim Georgia Institute of Technology
Wayne Li Georgia Institute of Technology