Optimizing machining time for CAD / CAM milling programming using the Taguchi method

The CNC milling machining process is a very important process in manufacturing products because it affects the profitability of the production process. CAD / CAM programming software tools speed up the process and predict the production process time. This study uses CAD / CAM simulation of HSM operations and uses the Taguchi method which is a method that can optimize process variables to get a good response. Process variables chosen are Speed, feed rate, depth of cut, step over, stay on surface and radius link. The type of tool used is an end mill with a diameter of 20 mm SECO Jabro-Solid-JS554 standard. The results of the study stated that the HSM machining simulation process that had a greater influence on machining time was feeding 680 mm / minute and 5 mm depth of cut. The experimental results confirm the ideal time produced is 34.25 minutes.


Introduction
Computer Aided Design (CAD) is one of the computer (or workstations) uses which has a function for assisting the creation, modification, analysis, or optimization of designs [1]. The CAD software is used to increase design productivity, improve design quality, improve communication through documentation, and to create databases for manufacturing [2]. Normally, CAD output is in the form of electronic files for printing, machining, or other manufacturing operations. Computer Aided Manufacturing (CAM) is the use of computer software to control machine tools or other machine parts related to machining processes. CAM is a process after modeling using Computer Aided Design (CAD) or also Computer Aided Engineering (CAE). CAM can also refer to the use of computers to assist in all plant operations, including planning, management, transportation, and storage [3,4].
The Taguchi design method was developed by Dr. Genichi Taguchi and is a set of methodologies in which the variability of the characteristics of the manufacturing process can be considered during the design phase itself [5]. The uniqueness of the Taguchi design methodology is that it can examine processes for a number of factors at once. In addition, it also considers the influence of noise factors other than the factors controlled. Julie Z. Zhang et al. [6] have examined optimizations to optimize surface roughness in milling operations using the Taguchi design and concluded that the Taguchi design is an efficient methodology for optimizing machining parameters [7].  [8] have studied the optimization of end milling parameters on hardened steel and also analyzed the effect of parameter interactions using the Taguchi design. W.H. Yang and YS. Tarng has used the Taguchi method to optimize cutting parameters in the replacement of S45C steel with tungsten carbide tools and together with the use of ANOVA to identify significant factors that affect operations [9]. Tsann-Rong Lin has carried out an experimental design to optimize cutting parameters from face grinding operations with TiN-coated devices [10].
Bagci Eyup and Şeref Aykut have identified optimal end surfaces in face surface grinding operations of cobalt-based alloys by optimizing milling parameters such as feed rate, cutting speed, and cutting depth [11]. CAD / CAM programming CNC (Computer Numerical Control) machines are used to make G-Code. CAD / CAM milling simulates the actual machining process. To produce minimal machining times, machining variables that affect the machining process can be set. The Taguchi experimental design method is used to obtain optimal machining times [12].

Method
The Taguchi method is a statistical method developed by Genichi Taguchi to improve the quality of manufacturing, engineering, biotechnology, marketing and advertising production. The advantage of the experimental design of the Taguchi method is getting an overview of how parameters affect process performance [5]. There are two main measuring tools in the Taguchi design method, namely: • Size of quality during design / development. In order to obtain a good individual indicator and can be used to evaluate due to changes in a parameter design, especially in product performance, the Signal-to-noise ratio (SNR) model is used.
• Efficient experiments to find information about the design of simultaneous parameters. From the experiments carried out must be able to obtain interrelated information so that design changes during the manufacture and use of customers can be avoided, and the information must be obtained in the minimum time and material. To perform an efficient experiment, the Orthogonal Matrix was used [13]. The following steps are performed to apply the Taguchi parameter design approach: • Determine the quality parameters to be optimized.
• Select significant controlled parameters and different levels for research.
• Select the appropriate orthogonal array.
• Do the number of experiments needed, simulate CNC milling programs from each level so that you get a response in the form of machining time. • Analysis of the results obtained from the experiment and determine the optimal level of the controlled parameters, calculation of the S / N ratio and the resulting graph. • The characteristics of the S / N ratio used are the smaller the better (Smaller is Better) where the equations are as follows [5]: • Predict optimized parameter performance.
• Confirm to get the actual results.
The main objective of this research paper is to apply the Taguchi parameter design to predict the main factors that influence the counter country milling operation and to optimize machining times.

Determining research factors and levels
The following are the factors and levels of research adapted to the Milling catalog (Jabro-Solid-JS554 SECO).  Orthogonal matrix is a matrix whose elements are arranged according to rows and columns (5). The selected factors are 6, namely speed, feed, depth of cut (ap), step over (ae), Stay on surface (SoS) and lead radius (Lr), while the research level is 5, then the orthogonal matrix L25 is chosen (56). The form of variations in levels can be seen in Table 2.

Data collection and S / N calculation of HSM ratio
There are two stages of data retrieval:   In Figure 2 it can be seen that the highest delta value is feed speed and feed depth (ap) so that it can be concluded that the two factors most influence the machining time. While the Smaller is Better target value is the average value of the S / N ratio close to zero which results in each factor. Then the variations in levels that produce machining processes with ideal time are: Speed (N1) 800 rpm, Feed (f5) 680 mm / minute, Feed depth (ap5) 5 mm, Step over (ae5) 16 mm, Stay on surface (SoS3) 30 mm, Link radius (Lr1) 2 mm. From Table 3. in graphical form as shown in figure 2 below.  Table 4 below is the confirmation of the HSM operational machining simulation time. Number 1 and 2 are two variations of the optimum machining time level while number 3 is confirmation of the optimum machining time for 34.25 minutes.  figure 3. the following.

Discussion
CAD/CAM programming simulation strategies can be made convenient for design and manufacturing problems, and provide an easier way to solve complex manufacturing problems. This reduces the time needed to apply to new CNC machines. Simulation can prevent the danger faced when dealing with CAM steps and CNC machines have accidents and near misses, which may occur at an early stage.

Conclusion
From the results of this study of machining parameters for machining time on CAD / CAM programming, HSM Milling can be concluded as follows: • In the machining simulation process with operational CAD / CAM software HSM the main factors that have greater influence are feeding and depth of cut, while Speed, stay on surface and radius links are less influential. Then the ideal combination of machining simulation parameters is: Speed (N1) 800 rpm, Feed (f5) 680 mm / minute, Feed depth (ap5) 5 mm, Step over (ae5) 16 mm, Stay on the surface (SoS3) 30 mm, Link radius (Lr1) 2 mm. • The results of the confirmation experiment produce the ideal machining simulation time by comparing the S / N ratio of the HSM machining simulation process. We could obtain the ideal time for 34.25 minutes.