Multi-objective optimization and prediction of surface roughness and printing time in FFF printed ABS polymer

Arivazhagan Selvam; Suresh Mayilswamy; Ruban Whenish; K Naresh; Vigneshwaran Shanmugam; Oisik Das

doi:10.1038/s41598-022-20782-8

Multi-objective optimization and prediction of surface roughness and printing time in FFF printed ABS polymer

Sci Rep. 2022 Oct 7;12(1):16887. doi: 10.1038/s41598-022-20782-8.

Authors

Arivazhagan Selvam¹, Suresh Mayilswamy², Ruban Whenish³, K Naresh⁴, Vigneshwaran Shanmugam^{5

6}, Oisik Das⁷

Affiliations

¹ Department of Mechanical Engineering, KPR Institute of Engineering and Technology, Coimbatore, Tamil Nadu, India. arivuazhagan001@gmail.com.
² Department of Robotics and Automation Engineering, PSG College of Technology, Coimbatore, Tamil Nadu, India.
³ Centre for Bio Materials, Cellular and Molecular Theranostics, Vellore Institute of Technology, Vellore, Tamil Nadu, India.
⁴ Department of Chemical Engineering and Materials Science, University of Southern California, Los Angeles, 90089, USA.
⁵ Department of Civil, Environmental and Natural Resources Engineering, Luleå University of Technology, 97187, Luleå, Sweden. s.vigneshwaren@gmail.com.
⁶ Department of Mechanical Engineering, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Chennai, India. s.vigneshwaren@gmail.com.
⁷ Department of Civil, Environmental and Natural Resources Engineering, Luleå University of Technology, 97187, Luleå, Sweden. oisik.das@ltu.se.

Abstract

In this study, fused filament fabrication (FFF) printing parameters were optimized to improve the surface quality and reduce the printing time of Acrylonitrile Butadiene Styrene (ABS) polymer using the Analysis of Variance (ANOVA), it is a statistical analysis tool. A multi-objective optimization technique was employed to predict the optimum process parameter values using particle swarm optimization (PSO) and response surface methodology (RSM) techniques. Printing time and surface roughness were analyzed as a function of layer thickness, printing speed and nozzle temperature. A central composite design was preferred by employing the RSM method, and experiments were carried out as per the design of experiments (DoE). To understand the relationship between the identified input parameters and the output responses, several mathematical models were developed. After validating the accuracy of the developed regression model, these models were then coupled with PSO and RSM to predict the optimum parameter values. Moreover, the weighted aggregated sum product assessment (WASPAS) ranking method was employed to compare the RSM and PSO to identify the best optimization technique. WASPAS ranking method shows PSO has finer optimal values [printing speed of 125.6 mm/sec, nozzle temperature of 221 °C and layer thickness of 0.29 mm] than the RSM method. The optimum values were compared with the experimental results. Predicted parameter values through the PSO method showed high surface quality for the type of the surfaces, i.e., the surface roughness value of flat upper and down surfaces is approximately 3.92 µm, and this value for the other surfaces is lower, which is approximately 1.78 µm, at a minimum printing time of 24 min.