This work aims to implement and use machine learning algorithms to predict the yield of bio-oil during the pyrolysis of lignocellulosic biomass based on the physicochemical properties and composition of the biomass feed and pyrolysis conditions. The biomass pyrolysis process is influenced by different process parameters, such as pyrolysis temperature, heating rate, composition of biomass, and purge gas flow rate. The inter-relation between the yield of different pyrolysis products and process parameters can be well predicted by using different machine learning algorithms. In this study, different machine learning algorithms, namely, multi-linear regression, gradient boosting, random forest, and decision tree, have been trained on the dataset and the models are compared to identify the optimum method for the determination of bio-oil yield prediction model. Analysis of the results showed the gradient boosting method to possess a regression score of 0.97 and 0.89 for the training and testing sets with root-mean-squared error (RMSE) values of 1.19 and 2.39, respectively, and overcome the problem of overfitting. Therefore, the present study provides an approach to train a generalized machine learning model, which can be employed on large datasets while avoiding the error of overfitting.

The demand for additive manufacturing (AM) continues to grow as more industries look to integrate the technology into their product development. However, there is a deficit of designers skilled to innovate with this technology due to challenges in supporting designers with tools and education for their development in design for AM (DfAM). There is a need to introduce intuitive tools and knowledge to enable future designers to DfAM. Immersive virtual reality (VR) shows promise to serve as an intuitive tool for DfAM to aid designers during design evaluation. The goal of this research is to, therefore, identify the effects of immersion in design evaluation and study how evaluating designs for DfAM between mediums that vary in immersion, affects the results of the DfAM evaluation and the mental effort experienced from evaluating the designs. Our findings suggest that designers can use immersive and non-immersive mediums for DfAM evaluation without experiencing significant differences in the outcomes of the evaluation and the cognitive load experienced from conducting the evaluation. The findings from this work thus have implications for how industries can customize product and designer-talent development using modular design evaluation systems that leverage capabilities in immersive and non-immersive DfAM evaluation.

Additive manufacturing (AM) is a rapidly growing technology within the industry and education sectors. Despite this, there lacks a comprehensive tool to guide AM-novices in evaluating the suitability of a given design for fabrication by the range of AM processes. Existing design for additive manufacturing (DfAM) evaluation tools tend to focus on only certain key process-dependent DfAM considerations. By contrast the purpose of this research is to propose a tool that guides a user to comprehensively evaluate their chosen design and educates the user on an appropriate DfAM strategy. The tool incorporates both opportunistic and restrictive elements, integrates the seven major AM processes and outputs an evaluative score and recommends processes and improvements for the input design. The paper presents a thorough framework for this evaluation tool and details the inclusion of features such as dual-DfAM consideration, process recommendations, and a weighting system for restrictive DfAM. The result is a detailed recommendation output that helps users to determine not only “can you print your design” but also “should you print your design” by combining several key research studies to build a comprehensive user design tool. This research demonstrates the potential of the framework through a series of case studies geometries. The preliminary framework presented in this paper establishes a foundation for future studies to refine the tool’s accuracy using more data and expert analysis.

This paper proposes an augmented reality (AR) framework and tool on smartphones as an alternative to conventional inspection for AM parts. The framework attempts to introduce the rapid inspection potential of smartphone based AR within manufacturing by leveraging the manufacturing capability of additive manufacturing (AM) to integrate markers onto AM parts. The key step from this framework that is explored in this paper is the design and quality assessment of AM markers for marker registration. As part of the marker design and quality assessment objectives, this research conducts an evaluation on the effects of different AM processes on the quality of augmentation achieved from AM fiducial markers. Furthermore, it evaluates the minimum fiducial pattern size that on integration onto AM parts will be viable for augmentation. The results suggest that the AM process and the size of the fiducial pattern play a significant role in determining the quality of the AM markers. The paper concludes by stating that dual material extrusion AM markers provide the highest number of detectable features and therefore the highest quality of AM markers, and the smallest viable fiducial pattern for Cybercode/QR code marker can be sized at 19 × 19mm2.

Although there is a substantial growth in the Additive Manufacturing (AM) market commensurate with the demand for products produced by AM methods, there is a shortage of skilled designers in the workforce that can apply AM effectively to meet this demand. This is due to the innate complications with cost and infrastructure for high-barrier-to-entry AM processes such as powder bed fusion when attempting to educate designers about these processes through in-person learning. To meet the demands for a skilled AM workforce while also accounting for the limited access to the range of AM processes, it is important to explore other mediums of AM education such as computer-aided instruction (CAI) which can increase access to hands-on learning experiences. Therefore, the purpose of this paper is to analyze the use of CAI in AM process education and focus on its effects on knowledge gain and cognitive load. Our findings show that when designers are educated about material extrusion and powder bed fusion through CAI, the knowledge gain for powder bed fusion is significantly different than knowledge gain for material extrusion, with no significant difference in cognitive load between these two AM processes. These findings imply that there is potential in virtual mediums to improve a designer’s process-centric knowledge for the full range of AM processes including those that are usually inaccessible. We take these findings to begin developing recommendations and guidelines for the use of virtual mediums in AM education and future research that investigates implications for virtual AM education.