Design Process for Additive Manufacturing

1. Introduction

Additive Manufacturing (AM) techniques are rapidly emerging as leading technologies for the creation of complex models [1,2]. The AM process involves sequentially depositing material in layers until the complete model is achieved. The time required to produce finished models using additive techniques can vary significantly, ranging from several hours to several days, depending on factors such as the manufacturing technology used, the object’s dimensions, and the complexity of its design [3]. Currently, there is a diverse range of AM techniques available in the market. In partnership with the International Organization for Standardization (ISO) and the American Society for Testing and Materials (ASTM), several standards, such as ISO/ASTM 52900 [4] and ISO/ASTM 52910 [5], have been established. These standards provide comprehensive descriptions of the additive techniques presently in use.

AM techniques are widely applied, including in the aerospace [6], automotive [7], and medical industries [8]. Since functional models are often produced using additive technologies, they must meet the design specifications. Unfortunately, there is still a lack of standardized design guidelines concerning areas such as mechanical strength assessments, geometrical tolerancing, and surface roughness. This lack of standards complicates the commercialization of finished products manufactured using these methods. The challenges in commercializing manufactured products are affected by the methods and parameters used during the design and AM phases. In the case of Computer-Aided Design (CAD) and Reverse Engineering (RE) design for AM, mechanical strength, geometric accuracy, and surface roughness are affected by the topology and geometry of a 3D-CAD model [9], CAD/RE software algorithms [10], chord and angular tolerance [11], mesh size, and scanning quality and data processing parameters in RE process [12]. The properties of a finished part produced through AM are influenced by a variety of factors. The mechanical strength of 3D-printed parts depends significantly on the type of material [13], the density and pattern of the internal infill in a model [14], number of shells [14], print direction [15], and proper nozzle [16] and bed temperatures [17], along with appropriate material cooling. The parameters that have the greatest impact on geometric accuracy and surface roughness include layer height [18], print speed [18], printer calibration [19], material shrinkage [19], print temperature and cooling [20], and print direction [21].

To achieve optimal mechanical strength, geometric accuracy, and surface roughness in the finished parts, a comprehensive approach is necessary. Conducting detailed scientific research is essential to optimize both the design process (CAD/RE) and the AM process. These two processes should not be treated separately; instead, they must be synchronized appropriately to ensure the highest quality, precision, and functionality of 3D prints.

2. An Overview of Published Articles

When designing a 3D CAD model for AM, it is crucial to ensure that the geometric parameters align with the most favorable operating conditions, such as tightness, accuracy, component connections, wear, and deformation, among others. This task can be challenging, as each AM technology has its own technical limitations, often causing the final manufactured model to differ significantly from the designer’s original intentions. Therefore, optimizing the geometry during the CAD modeling stage is essential to make it suitable for AM. Hence, the article by Metzger et al. (contribution 1) presents a new, user-friendly calculation model for the design of thin-walled, slit-shaped reactor modules created using AM. The simplified calculation model proved reliable, and a design graph was created based on the findings. Engineers can use this graph to design safe and robust additively manufactured reactor modules efficiently. The study highlights the potential of AM for creating high-performance process equipment with minimal material usage.

Geometry optimization is also undertaken in relation to the developed CAD models of prostheses. The authors of contribution 2 conducted a Finite Element Analysis (FEA) to compare two forelimb prosthetic designs for dogs: a solid model and a perforated model. Both models, made from ABS plastic, were tested under a simulated static load equivalent to 60% of a dog’s body weight. The analysis revealed that both models remained within the material’s elastic limit, indicating they would not experience permanent deformation under the applied loads. As a result of these findings, the perforated model was determined to be the more suitable option for developing canine prosthetics. Future work will involve analyzing the designs under dynamic loads, such as those encountered during walking and running.

Currently, research is also being conducted on optimizing geometry for AM using Artificial Intelligence (AI). The article by Zichar et al. (contribution 3) explores how AI tools, such as ChatGPT and Gemini, can assist or even replace students in code-based 3D modeling tasks. The study found that students are not yet able to delegate their work to AI tools fully. While AI can generate code for simple models, it often requires modifications and corrections to meet specific requirements. Furthermore, the article highlights that the increasing popularity of AI necessitates a reevaluation of assessment methods by educators. Instead of fearing that AI will take over students’ work, teachers should focus on understanding how these tools can support, rather than replace, the learning process.

In the early stages of design, engineers often lack the necessary knowledge and tools to make informed decisions about manufacturing technologies. This can result in inefficient and costly choices. Although AM provides many advantages, it is not always the best solution for every application. The article by Salmi et al. (contribution 4) describes a hybrid Multi-Criteria Decision-Making (MCDM) method that aids engineers in selecting between AM and conventional manufacturing processes (such as machining) for specific components. This method analyzes several parameters to deliver a thorough assessment, including

• Part geometry: shape complexity and wall thickness;
• Material requirements: type of material and its properties;
• Production considerations: required accuracy, tolerances, and the number of parts to be produced;
• Economic factors: material and process costs, along with production time;
• Environmental factors: energy consumption and material waste.

This decision-making model is intended for use during the preliminary design phase.

When creating a 3D CAD model for AM, traditional CAD modeling techniques are commonly used. However, challenges often arise when there is a lack of technological or material documentation for a product. This situation is particularly prevalent when designing models of anatomical structures, museum artifacts, or other complex geometric shapes, where solid or surface design may not be feasible. The RE process can address these issues thanks to advancements in coordinate measuring systems, data processing software, and modern manufacturing techniques. Although this design process is frequently utilized for developing 3D CAD models intended for AM, it can sometimes result in geometric mapping errors during the design phase. Therefore, it is essential to establish procedures at the geometry design stage of the RE process to minimize these errors. The reconstruction process can be applied to different types of geometries. The article by Turek et al. titled (contribution 5) explores the errors that occur during RE and AM for models with simple geometries, including axisymmetric and primitive shapes. The research findings indicated that, overall, 95% of the points representing reconstruction errors are within the maximum deviation range of ±0.6 mm to ±1 mm. The highest errors in CAD modeling were attributed to the auto-surfacing method; overall, 95% of the points are within the average range of ±0.9 mm. In contrast, the smallest errors, averaging ±0.6 mm, occurred with the detect primitives method. Overall, on average, 95% of the points representing the surface of a model made using the additive manufacturing technology fall within the deviation range ±0.2 mm. The findings provide crucial insights for designers utilizing RE and AM techniques to create functional model replicas.

AM may also be used for models used in medicine. The article by Simarmata et al. (contribution 6) discusses a methodology for designing and producing personalized orthopedic insoles for patients with flat feet. The authors used an integrated design approach that combines several methods to create customized insoles that effectively reduce foot pressure. The analysis revealed that insoles 3D printed with a 20% auxetic infill made from TPU filament best aligned with patient preferences for both functionality and comfort. Measurements of pressure distribution confirmed that the insoles with the auxetic structure reduced maximum pressure by 25.4% compared to not using insoles, demonstrating their efficacy in distributing pressure across the foot. The article by Turek et al. (contribution 7) focuses on enhancing the accuracy of AM for medical applications, particularly in the reconstruction of the zygomatic bone. The authors explored the use of Masked Stereolithography (mSLA) technology as a production method for these models due to its cost-effectiveness and high precision. They tested two printing modes, standard and ultralight, on an Anycubic Photon M3 Premium printer to determine which mode yielded superior geometric accuracy. The models were then validated using a structured light scanner. The results indicated that the ultralight mode offered better surface accuracy, which is crucial for planning precise surgical procedures. Notably, over 70% of the models’ surfaces fell within a deviation range of ±0.3 mm. This study demonstrates that when configured correctly, mSLA technology can effectively produce highly accurate surgical templates and implant-forming tools. This advancement significantly supports craniomaxillofacial reconstruction procedures.

The RE process and 3D printing can be used to create terrain models. The article by Chlost et al. (contribution 8) outlines a methodology for creating a 3D terrain model from point cloud data. This study aims to reduce mesh surface errors and incorporate a smoothing factor. The initial surface was generated from a square grid and subsequently converted to an input format suitable for a CAD environment. To minimize surface defects, a bilinear interpolation algorithm was applied. Accuracy analyses of the terrain mapping were conducted on three samples with different geometries using two options in Siemens NX software: Uniform Density (UD) and Variable Density (VD). The results indicated that changing the smoothing factor from 0% to 15% did not significantly affect accuracy. However, a marked increase in inaccuracy was noted at a smoothing factor of 20%. The developed methodology ensures high-accuracy mapping of digital data, which can be leveraged in manufacturing processes like 3D printing. Based on the results, three prints were produced using the Fused Deposition Modeling (FDM) method, each representing one of the analyzed terrain geometries.

The AM process is quite complex, and various 3D printing parameters, such as temperature, speed, and model orientation, directly influence the microstructure and mechanical properties of the material. Additionally, optimizing both the 3D printer design and the materials used in AM can impact issues like porosity, layer delamination, and internal stresses. These factors can significantly reduce the strength and dimensional accuracy of the final product. The article by Bradshaw et al. (contribution 9) is a review that examines how the unique layer-by-layer process of 3D Concrete Printing (3DPC) affects the long-term durability of the material. It outlines key challenges and recent advancements in mix design and testing methods. The review emphasizes that the durability of 3DPC can be significantly enhanced by optimizing both the mix design and the printing process. Additionally, it stresses the need for standardized testing protocols to accurately assess durability, particularly for issues such as freeze–thaw resistance, chloride ingress, and carbonation, which are particularly sensitive to the printing parameters and curing conditions. The article by Woods et al. (contribution 10) presents a novel design for a 3D-bioprinting printhead that enhances the precision and stability of the printing process. Many current bioprinters face challenges in accurately dispensing bio-inks, which are temperature-sensitive and contain living cells. Variations in the dispensing process or temperature can result in cell damage and decreased print quality. The authors propose a new printhead design that combines two essential control systems: pneumatic and thermal control. By integrating these systems into a modified extrusion bioprinter, they enable accurate and reliable dispensing of bio-inks, which is critical for tissue engineering. This study highlights how engineering innovations in hardware can advance bioprinting technology, facilitating the creation of more precise and functional tissue structures. Researching the implementation of 4D printing is currently essential. Four-dimensiaonal printing is an extension of traditional 3D printing. It involves the fabrication of objects using innovative materials that are capable of changing their shape, properties, or functionality over time in response to an external stimulus. The article by Jin et al. (contribution 11) defines the main areas of application:

• Biomedicine: smart implants, self-degrading tissue scaffolds, and drug delivery systems;
• Robotics: soft robots that can change shape and move without complex motors;
• Aerospace: components that can alter their shape in response to environmental conditions, such as aircraft wings;
• Civil engineering: self-healing structures.

The article concludes that 4D printing is still in its early stages, and the most significant challenges remain high costs, the limited stability of printed structures, and the need for further research into new materials.

3. Conclusions

The articles in the Special Issue discussed essential aspects of AM. They emphasize that this process, which involves the layer-by-layer deposition of material, requires optimization during both the design and manufacturing stages to achieve the desired properties, such as strength, accuracy, and surface quality. The quality of a finished 3D-printed part is influenced by a combination of complex factors that depend on the specific technology and machine settings used. These factors can be categorized into two main groups:

• Design parameters (CAD/RE): these include the model’s topology, geometry, wall thickness, and the presence of features like sharp angles or hollow spaces. The algorithms used for triangulation and mesh optimization, as well as the chord and angular tolerances from the CAD to STL export process, also directly affect the final part’s surface quality. In RE, the quality of the 3D scanner and data processing parameters are critical for accuracy;
• Manufacturing parameters: these factors directly impact the part’s mechanical strength and geometric accuracy. They include the type of material used (e.g., PLA, ABS, nylon), infill density and pattern, the number of perimeters or “shells,” print direction, and temperature and cooling settings. For geometric accuracy and surface roughness, the key parameters are layer height, print speed, printer calibration, and material shrinkage. The articles note that a smaller layer height improves precision but increases print time.

The articles also highlight several studies that demonstrate the importance of this optimization:

• Engineering and design: research on a simplified design method for thin-walled reactor modules allowed for the creation of safe and robust components with minimal material use. In another example, an FEA of a perforated prosthetic model for dogs was found to be more suitable than a solid model, highlighting the benefits of geometry optimization;
• Medical applications: studies show that AM is used to create personalized orthopedic insoles for flat-footed patients, where an auxetic infill reduced maximum foot pressure by 25.4%. Another study focused on using mSLA for surgical templates in zygomatic bone reconstruction, finding that an “Ultralight” printing mode provided the superior surface accuracy essential for surgical planning.
• Reverse engineering: reverse engineering is useful for creating models with complex shapes, which can lead to geometric mapping errors. One publication found that the highest errors occurred with the “auto-surfacing” method in CAD modeling.

The articles conclude that while standards like ISO/ASTM 52900 exist, there is still a lack of standardized design guidelines for critical areas like mechanical strength and dimensional tolerancing. Therefore, a comprehensive approach that links both design and manufacturing processes is essential to ensure high-quality and functional 3D prints.