Occupation Clustering Methodology for Training In-Demand Engineering Middle-Skilled Workers in the Advanced Manufacturing Industry

The manufacturing industry faces a labor short-age and a mismatch between available and desired em-ployee skills. This challenge is more acute for advanced manufacturing and for middle-skilled engineering work-ers. Despite this need, funding to develop middle-skilled training programs is severely limited. This article addresses this challenge by providing a method for identifying high-value, shared skills to maximize training resources. A novel set of methods were developed to classify advanced manufacturing industries, quantitatively characterize en-gineering skills gaps, and identify occupations with shared in-demand skills. These methods were applied to a dataset collected across five industries: robotics, flexible electron-ics, advanced fabrics and fibers, integrated photonics, and 3D/additive manufacturing. Semi-structured interviews with industry experts were performed to quantify skill importance and skill gaps. A combination of rank-biased overlap analysis and unsupervised clustering methods were used to identify occupations with shared in-demand skills. Results suggested that the eight most prevalent middle-skilled occupations could be trained effectively in four programs by sharing training resources. These train-ing recommendations can prepare tomorrow's engineering technology workforce while maximizing scarce curricu-lum development resources.