Smart composite-based additive manufacturing with functional compression machine learning for remanufacturing recycled heterogeneous fibers

Arnold William Bangel; Amanda Nitta; Nazanin Tabatabaei; Chao Wang; Xuan Song

doi:10.1016/j.jmapro.2026.06.028

Back

Smart composite-based additive manufacturing with functional compression machine learning for remanufacturing recycled heterogeneous fibers

Journal article

Peer reviewed

Smart composite-based additive manufacturing with functional compression machine learning for remanufacturing recycled heterogeneous fibers

Arnold William Bangel, Amanda Nitta, Nazanin Tabatabaei, Chao Wang and Xuan Song

Journal of manufacturing processes, Vol.172, pp.1289-1301

08/30/2026

DOI: 10.1016/j.jmapro.2026.06.028

View Online

Abstract

Using recycled materials, such as fibers, as feedstocks for additive manufacturing (AM) has attracted significant interest as a potential route to extend material lifecycles and reduce production costs. However, the intrinsic heterogeneity of recycled materials presents substantial challenges for AM processing, as most AM techniques require stringent control over feedstock purity and consistency. In this work, we propose a smart AM framework that integrates in situ optical sensing with functional data compression to dynamically manage highly heterogeneous fiber feedstocks in a composite-based additive manufacturing (CBAM) platform. The framework leverages data acquired from systematic testing of diverse feedstocks under varying process parameters to dynamically optimize processing conditions for a given feedstock and feeds this information back into the printing process in real time to enable adaptive process adjustments. By combining with advanced feedstock delivery systems, the approach enables the fabrication of parts with improved accuracy and mechanical performance. We demonstrate that variations in fiber properties strongly influence key processing parameters that govern the strength of the manufactured composite components. Implementation of the proposed smart CBAM framework results in a strength increase of nearly 120%.

Machine Learning

Composites

Optical imaging

Recycled fibers

Smart additive manufacturing

Details

Title: Subtitle: Smart composite-based additive manufacturing with functional compression machine learning for remanufacturing recycled heterogeneous fibers
Creators: Arnold William Bangel - University of Iowa
Amanda Nitta - University of Iowa
Nazanin Tabatabaei - University of Iowa
Chao Wang - University of Iowa
Xuan Song - University of Iowa
Resource Type: Journal article
Publication Details: Journal of manufacturing processes, Vol.172, pp.1289-1301
DOI: 10.1016/j.jmapro.2026.06.028
ISSN: 1526-6125
eISSN: 2212-4616
Publisher: Elsevier Ltd
Number of pages: 13
Grant note: U.S. Department of Energy's Office of Energy Efficiency and Renewable Energy (EERE) under the Advanced Manufacturing Office: DE-EE0007897 U.S. National Science Foundation: 2242763 U.S. National Science Foundation: 2500269
This material is based upon work supported by the U.S. Department of Energy's Office of Energy Efficiency and Renewable Energy (EERE) under the Advanced Manufacturing Office Award Number DE-EE0007897 awarded to the REMADE Institute, a division of Sustainable Manufacturing Innovation Alliance Corp. A.B., N.T., and X.S. also acknowledge the support from the U.S. National Science Foundation (Award No. 2242763). A.N. and C.W. also acknowledge the support from the U.S. National Science Foundation (Award No. 2500269). We thank undergraduate student Diego Robles for assisting with experiments and graduated PhD student Amirhossein Fallahdizcheh for providing imaging processing algorithms.
Language: English
Electronic publication date: 06/26/2026
Date published: 08/30/2026
Academic Unit: Industrial and Systems Engineering; Injury Prevention Research Center; Mechanical Engineering
Record Identifier: 9985179094002771

Metrics

1 Record Views