*1.1. Additive Manufacturing*

Currently, the manufacturing industry is a sector highly globalized with a constant need for productivity gains and innovation. In this regard, additive manufacturing (AM) is considered to be one of the latest manufacturing revolutions and a future leading edge technology [1]. Additive manufacturing is entering the market to meet the demand of custom parts of complex geometry and reduce investment in tooling. Nowadays, this manufacturing process is still considered as a promising technology and is studied extensively in order to assess its viability in commercial applications such as electronics (resistors and sensors), optical (antennas), medical (artificial hip joints, bone structures, and tissue scaffolds), automotive, communication, and aerospace industries (engines, turbines, and thermal insulation coatings) [2]. Despite the great improvements that have been made in recent years, additive manufacturing still has some limitations. For instance, Oropalloand Piegl [3] identified ten challenges that should be conveniently studied and solved in coming years, such as shape optimization, design for 3D printing, or pre- and post-processing.

Additive manufacturing is characterized by the manufacture of pieces from a CAD model through the accumulation and joining of layers for obtaining the desired physical model. Recently,

ASTM International defined a body of terms for additive manufacturing [4]. The different types of processes can be classified depending on [5]: (a) raw material (liquid, powder or solid); and (b) the kind of physical joint between the material. Currently, there are available several processes such as Stereolithography (SLA) [6], Selective Laser Sintering/Melting (SLS/SLM) [7], Laminated Object Manufacturing (LOM) [8], and Fused Deposition Modeling (FDM) [9,10].

The selection of the additive manufacturing process must take into account the pros and cons of each of the technologies. For instance, the FDM process is simple, which makes it a suitable candidate for being chosen by general users. Its main advantages are [10–13]: low machinery cost, no expensive tooling is necessary, broad range of materials, high durability of the components, acceptable dimensional accuracy, and not being time consuming. But, the process also has some disadvantages, such as low mechanical strength, difficulty to obtain thin walls, and poor surface quality. Polymers are widely used as the main base material in FDM processes. Typical materials include PLA [14] and ABS [15], but composite materials are also being adopted for manufacturing complex components. For instance, Fe-nylon6 composite wires were compared to ABS solutions, concluding that the composite materials are highly wear resistant [16].Polypropylene reinforced with glass-fiber wasstudied, showing adequate mechanical properties for small series of parts [17].

Although FDM processes have significant industrial value for manufacturing complex components, there is a need tocarry out proper research focused on prominent aspects such as surface roughness and performance optimization. The performance of the manufactured parts depends upon a large number of process factors, such as the type of material and process parameters, so it is quite difficult to obtain an ideal FDM process that fulfils all the requirements, particularly producing products of high surface quality.

#### *1.2. Surface Quality in FDM Processes*

In additive manufacturing, in general, pre-processing and post-processing activities should be carried out [3]. However, the quality of the parts is not adequate when compared to other mature manufacturing processes, such as machining. One of the main problems for obtaining good surface quality in additive manufacturing is the staircase effect. According to Strano et al. [18], usually manual post-processing operations are needed for obtaining adequate surface roughness because complex geometries compromise the advantages of additive manufacturing. Pandey et al. [19] analyzed the staircase effect that generates "chordal error" between an original surface of a CAD model and the corresponding triangle in the tessellated model. The authors concluded that the tessellation and slicing during the manufacturing process are two sources of surface inaccuracies that must be taken into account.

Various studies have been specifically carried out on FDM process parameters, discussing their effect on outputs, such as mechanical properties, and surface topography and quality [20–22]. For instance, Altan et al. [14] studied the effect of process parameters on surface roughness and the tensile strength on polylactic acid (PLA) samples. The samples were fabricated as per the ASTM standards and a Taguchi L16 experimental design, using three parameters: layer thickness, deposition head velocity, and nozzle temperature. The authors concluded that the layer thickness and deposition head velocity are dominant factors on surface roughness.

Campbell et al. [23] investigated surface roughness for different materials. The authors found that, in the case of ABS material, when using layer thickness of 0.253 mm, the surface roughness values for FDM processes ranged between 9 μm and 40 μm. Recently, Akande et al. [24] analyzed the optimal process parameters for obtaining good surface finish and dimensional accuracy. The authors employed a layer height of 0.25 and 0.5 mm, varying the filling density and speed of deposition, identifying that the surface roughness for PLA material ranged between 2.46 μm and 22.48 μm. Altan et al. [14] used layer thickness between 0.1 mm and 0.4 mm to create PLA samples using a FDM process. The surface roughness obtained varied within the range of 9.102 to 10.275 μm.

From previous scientific records, it has been identified that the performance of the FDM process extensively depends upon its process parameters and their levels. However, the number of publications dealing with the identification of critical factors and the optimization of the manufacturing process depending on adequate selection of factors and levels is still limited, particularly when it comes to surface roughness.

The optimization of the process parameters used for printing is an adequate strategy for improving part performance in terms of surface finish. So, the present paper addresses the study of surface roughness for FDM pieces, which has not been studied in detail in the literature. The paper shows an experimental study on fused deposition modeling, analyzing the quality of the parts after varying a set of printing parameters: layer height, wall thickness, printing speed, temperature, and printing path using both statistical and graphical methods.
