Quantifying Green Manufacturability of a Unit Production Process Using Simulation

Available online at www.sciencedirect.com

ScienceDirect Procedia CIRP 29 (2015) 257 – 262

The 22nd CIRP conference on Life Cycle Engineering

Quantifying green manufacturability of a unit production process using simulation Amandeep Singha,*, Deepu Philipb, J. Ramkumarc a Doctoral Student, Industrial and Management Engineering Department, Indian Institute of Technology, Kanpur-208016, India Assistant Professor, Industrial and Management Engineering Department, Indian Institute of Technology, Kanpur-208016, India c Associate Professor, Mechanical Engineering Department, Indian Institute of Technology, Kanpur-208016, India *Amandeep Singh . Tel.: +091-904-461-4643; fax: +91-512-259-7553. E-mail address: [email protected] b

Abstract Consumer awareness towards environment and sustainability is on the rise, which is forcing industry to face the challenge of balancing economic and monetary priorities against environmental and social accountabilities. New methods and techniques are being developed to cater these challenges in an efficient way. This work aims to develop a roadmap to convert a unit manufacturing process more ‘green’ by minimizing resource utilization. Live laboratory experiments were conducted to model various factors influencing the ‘greenness’ of the unit process which was surface grinding. We evaluated various control strategies (factor settings) using ABC algorithm, implemented in a numerical simulation mode. The experimental data suggested significant improvement on the Green Index of the process. © The Authors. Authors. Published Published by by Elsevier B.V. This is an open access article under the CC BY-NC-ND license © 2015 2015 The Elsevier B.V. Peer-review under responsibility of the International Scientific Committee of the Conference “22nd CIRP conference on Life Cycle (http://creativecommons.org/licenses/by-nc-nd/4.0/). Engineering. Peer-review under responsibility of the scientific committee of The 22nd CIRP conference on Life Cycle Engineering Keywords: Green manufacturing; simulation; ABC algorithm; surface grinding

1. Introduction Manufacturing is the engine that drives industrialized advancement, and sustainability is one of the most important recent manufacturing considerations. Researchers and industry consider sustainability of manufacturing processes as a vital challenge. Nomenclature ABC DOC Ft Fn Ra Vw

Artificial bee colony Depth of cut (μm) Tangential cutting force (N) Normal cutting force (N) Roughness parameter (μm) Table speed (m/min)

Glavic and Lukman [1] define sustainable production as “creating goods by using processes and systems that are nonpolluting, that conserve energy and natural resources in

economically viable, safe and healthy ways for employees, communities, and consumers and which are socially and creatively rewarding for all stakeholders for the short and long-term future”. 1.1. Green Manufacturing Green Manufacturing is amongst the key elements that fall under the overarching umbrella of sustainability. Green manufacturing deals with maintaining environmental, economical, and social objectives of sustainability in the manufacturing domain. Some examples of sustainable green manufacturing activities are, reducing hazardous emissions, eliminating wasteful resources consumption, and recycling [2]. Manufacturing is the major segment for industrial energy consumption, and industrial pollution. It is accountable for 84% of energy-related industry CO2 emissions, and also consumes 90% of industry energy [3].

2212-8271 © 2015 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the scientific committee of The 22nd CIRP conference on Life Cycle Engineering doi:10.1016/j.procir.2015.01.034

258

Amandeep Singh et al. / Procedia CIRP 29 (2015) 257 – 262

1.1.1 Greenness of a product

1.3. Green manufacturing in grinding process

A product can be said to be green in terms of one of the three types of environmental impacts [4]. That is, if environmental impact is:

In recent years there has been considerable research in green manufacturing in the field of grinding processes.

x x x

1.3.1 Green aspects in grinding process lower than conventional products, or the impact is null, or it has positive contribution to environment,

From the Green Option Matrix developed by Dangelico, and Pierpaolo [4], examples may be: x

x

A green product that focuses on energy is a consumes less energy than product that conventional products, or if a part of the energy used comes from renewable energy A green product that focuses on pollution is a product that is less polluting than conventional products.

This work focuses on two factors of greenness: (i) reducing energy and (ii) reducing pollution. 1.1.2 Constraints of industry Manufacturers are highly concerned about enhancing the energy and resource efficiency. However, adaptation of some energy efficient technologies, which are established using techniques based on incorrect estimation, can have undesirable impacts on production [5]. Industry is apprehensive about certain constraints while evaluating the efficiency improving technologies [6]. Besides production constraints, and technology constraints, resource (financial) constraint is of utmost importance for the industry to implement any energy efficiency enhancing measures. 1.2. Unit manufacturing process From systems viewpoint of manufacturing, manufacturing activities are considered as being composed of multiple levels, from an elemental level (unit process) to that of the entire enterprise, comprising all the activities in the manufacturing system. Duflou et al. [3] define single machine tools as the smallest unit, of which production systems are composed of. Each unit process interacts with other unit processes, and the ecosphere through its boundaries [7]. Energy consumption at unit level in manufacturing processes is dominated mainly by two drivers: x the cutting time, and x the process efficiency in terms of cutting force, and torques [8]. We considered a unit manufacturing process as the focus of this research to present initial aspects of a broader framework for green manufacturing.

Among the machining processes, grinding requires remarkably higher energies to remove a specific material volume when compared to other conventional machining processes like turning, milling etc. The main reason for this is, the high amount of rubbing and ploughing actions happening because of large number of cutting edges, or abrasive grits [9]. Hashimoto [10], by studying the energy consumption for grinding using different chip thicknesses, and for hard turning the same type of workpieces found that abrasive processes like grinding result in higher energy consumption. Furthermore, water-based grinding fluid is widely used for lubrication to improve grinding efficiency, cooling, and cleaning. Constituents of conventional grinding fluid are bases of chlorine, sulphur, and phosphorus, which are harmful, and also cause environmental pollution. Also, usage of grinding fluids results in high costs. For example, in several automobile fabrication, 15–30% of the machining costs for an automobile manufacture is related to the use of grinding fluid [11]. Besides, the delivery and cleaning of fluids needs high amount of energy. It has been observed that under certain circumstances, up to 32% of the total energy usage of a manufacturing plant has been attributed to cutting fluids, and it is even higher when cooling related auxiliaries such as mist collectors are included [12]. Researchers have worked on alternative (ecologically benign) cutting fluid [1315], for example Winter et al. [13] developed an approach which was applied in a grinding case suggesting water miscible polymer dilution as better alternative over nonmiscible mixable grinding oil. Winter et al. [13] has defined manufacturing processes as a unit process (adapted from Li et al. [16], who also proposed an integrated approach to assess eco-efficiency of unit manufacturing process). We selected grinding as a ‘unit manufacturing process’ for this study and emphasized on the aforementioned facets of greenness using the unit process. 1.3.2 Minimum Quantity Lubrication in grinding process 

Minimum quantity lubrication (MQL) is a technique that uses a spray of small oil droplets in a compressed air jet. The lubricant is sprayed directly into the cutting zone, as a replacement for the huge flows of conventional flood coolant. Since the air jet carries the oil droplets directly in the cutting area, it provides efficient lubrication [17]. However special set up and arrangements are required when conventional flood coolant is used in order to allow the fluid to penetrate the cutting zone efficiently. Researchers have presented MQL as one of the prominent techniques for green manufacturing. For example, Sadeghi and Haddad [18], analysed using the experimental data that in grinding of Ti6Al4V, better grinding performance is given by MQL technique as compared to conventional lubrication methods.

259

Amandeep Singh et al. / Procedia CIRP 29 (2015) 257 – 262

MQL fluid application method is compared with other methods in this paper. 1.4. Simulation in manufacturing Many techniques that use simulation iterations have been sprouted for locating optimal solutions to intricate, and difficult optimization problems that emerged in the field of engineering, and management. A few among these are evolution Strategies Evolutionary Programming (EP) [19], Genetic Algorithms (GA) [20], Particle Swarm Optimization (PSO) [21], Differential Evolution (DE) [22], Ant Colony Optimization (ACO) [23], Bacterial Foraging Optimization Algorithm (BFOA) [24], Simulated Annealing (SA) [25], and Artificial Bee colony algorithm (ABC) [26][27]. An artificial bee colony (ABC) algorithm falls under the umbrella of swarm intelligent based algorithms. It involves the intelligence of the foraging behavior of swarm of bees to find solutions for optimization problems. ABC algorithm with its variants has been used successfully to solve various unconstrained numerical optimization problems. In the experiments, the extended version of ABC algorithm has shown better performance than Particle Swarm Optimization (PSO) and, Differential Evolution (DE) [26] [27]. ABC algorithm employing a greedy selection mechanism as the selection process is adapted from Samanta et al. [28]. ABC has found vast applications in machining domain also [28-34]. For example Rao and Pawar [29] applied ABC, harmony search, and SA algorithms for multi-objective optimization of machining parameters in a grinding process. A simulation implementation of ABC algorithm was used in this work to find the optimal solution to the model generated from the experimental data.

Fig. 1. Schematic diagram of measure system of grinding force

Separate nozzles and respective set ups were used for flood and MQL method of fluid application at toolworkpiece interface. Experiments were carried out using a horizontal surface grinding machine manufactured by Hindustan Machine Tools Ltd., Bangalore, India. As shown in Fig. 2.

2. Experimental work Fig. 2. Horizontal surface grinder machine

Unit manufacturing process selected here is surface grinding. Input variables selected are Depth of cut (DOC), and table speed (Vw) under three different cooling mediums viz-a-viz dry, flood and MQL. Output variables measured are tangential force (Ft), normal force (Fn), and surface roughness (Ra). The values of Ft, and Fn are measured by online monitoring system and surface roughness (Ra) is measured using Surtonic Surface Profiler after each experiment. Workpiece used for the experimentation is Titanium alloy (Ti6Al4V), that has wide application in aircraft manufacturing. Workpiece of dimensions, 160 mm × 25 mm × 5 mm is mounted over MS plate size, 179 mm × 60 mm × 16 mm, which is then mounted directly on to the dynamometer with the help of screws. MS plate is used to avoid thermal distortion of dynamometer. Aluminium abrasive grinding wheel with specifications A46J5V10 is used. The grinding forces were measured with the help of KISTLER Piezo-electric dynamometer, type 9257A, and the data is stored, and presented by using LabVIEW software, as per the application requirement. The schematic for grinding force measuring system is given in Fig.1 .

Full factorial experimental design for all combination of parameters values is considered for the present case, with 3 levels of depth of cut, and 3 levels of table speed under 3 ways (dry, flood and MQL) of fluid application. In total there are 27 (3x3x3) sets of experiments. Cutting conditions selected for this unit manufacturing process are given in Table 1. Table 1. Cutting conditions for unit manufacturing process Parameter

Value

Grinding wheel speed

20 m/s

Depth of cut (DOC)

6 μm, 10 μm and 14 μm

Table speed (Vw)

5 m/min, 10 m/min, and 15 m/min

Grinding wheel

A46J5V10

Coolant flow rate

4.5 litre/minute

MQL flow rate

80 ml/hr

Grinding wheel mounting, balancing, and alignment, nozzles flow rates, dynamometer output, surface roughness profiler were all calibrated before starting the experiments.

260

Amandeep Singh et al. / Procedia CIRP 29 (2015) 257 – 262

3. Results and discussions We considered the spindle (wheel rotation) energy as the energy consumed in the unit manufacturing process; which in turn is a function of the cutting force. Surface roughness was also considered as a quality measure that indicated any major drop in process parameters. The effect of cooling modes on two components of cutting forces and surface roughness are presented in Fig. 3, Fig. 4, and Fig. 5.

Table 3. ANOVA for Fn model using three factors Source DF Adj SS Adj MS Vw 2 650.6 325.30

Fig. 5. (a) Surface Roughness vs Depth of cut (Vw=10 m/min); (b) Surface Roughness vs Table speed (DOC=10 μm)

Analysis of Variance (ANOVA) was used to measure the factors influencing the greenness of the unit process. The ANOVA results are summarized in Tables 2, 3, and 4 for F t, Fn, and Ra respectively. F-Value 8.72

p-value 0.010

DOC

2

916.30

458.15

38.22

0.000

Medium

2

2172.14

1086.07

90.61

0.000

Vw×DOC

4

9.97

2.49

0.21

0.927

Vw×Medium

4

52.70

13.18

1.10

0.419

DOC×Medium

4

86.01

21.50

1.79

0.223

Error

8

95.89

11.99

Total

26

3541.95

0.000

2

4493.5

2246.75

98.31

Medium

2

3433.2

1716.58

75.11

0.000

Vw×DOC

4

230.0

57.49

2.52

0.124

Vw×Medium

4

347.5

86.88

3.80

0.051

DOC×Medium

4

122.3

30.58

1.34

0.335

Error

8

182.8

22.85

Total

26

9459.9

F-Value 19.82

p-value 0.001

DOC

2

0.17912

0.089561

23.52

0.000

Medium

2

0.93064

0.465319

122.18

0.000

Vw×DOC

4

0.00467

0.001168

0.31

0.866

Vw×Medium

4

0.12330

0.030826

8.09

0.006

DOC×Medium

4

0.01629

0.004072

1.07

0.432

Error

8

0.03047

0.003808

Total

26

1.43543

All three ANOVA tables demonstrate that medium of fluid application significantly influence the process (largest F-value). Also from Figures 3 to 5; it can be noted that MQL mode of cooling reduces cutting force, with similar roughness of that of flood cooling mode. Hence, it can be argued that MQL method of fluid application reduces the energy consumption significantly as against the dry, and flood cooling methods. Regression models were obtained for the dependent variables (Ft, Fn, and Ra) with Vw and DOC as independent variables, and ‘Medium’ as the categorical variable. Regression analysis was conducted using Minitab17 software. The resulting model consisted of three parallel equations for the three types of medium (namely dry, flood, and MQL). For example equations for Ft were:

Fig. 4. (a) Normal force vs depth of cut (Vw=10 m/min); (b) Normal force vs table speed (DOC=10 μm)

Table 2. ANOVA for Ft model using three factors Source DF Adj SS Adj MS Vw 2 208.94 104.47

p-value 0.002

DOC

Table 4. ANOVA for Ra model using three factors Source DF Adj SS Adj MS Vw 2 0.15094 0.075468 Fig. 3. (a) Tangential force vs depth of cut (Vw=10 m/min); (b) Tangential force vs table speed (DOC=10 μm)

F-Value 14.23

Dry,

Ft = 15.4 + 0.28.Vw + 5.84.DOC +0.0150.Vw.DOC + 0.0155.Vw2 - 0.2181 DOC2.

Flood, Ft = 0.9 + 0.28.Vw + 5.84.DOC + 0.0150.V w.DOC + 0.0155.Vw2 - 0.2181 DOC2. MQL,

Ft = -6.0 + 0.28.Vw + 5.84.DOC +0.0150.Vw.DOC + 0.0155.Vw2 - 0.2181 DOC2.

The MQL model provided the minimum intercept value, whereas, other regression coefficients were similar for all the three mediums. This can also be observed from Figures 3-5, where the green line (with triangle shape markers) depicts the MQL case, which is lower than all other mediums. Hence, regression equations (shown below) for Ft, Fn, and Ra for the MQL case were used for the optimization stage.

261

Amandeep Singh et al. / Procedia CIRP 29 (2015) 257 – 262

Ft = -6.0 + 0.28.Vw + 5.84.DOC + 0.0150.Vw.DOC + 0.0155.Vw2 - 0.2181 DOC2. Fn= -32.4+0.88.Vw + 14.86.DOC + 0.1159.Vw.DOC - 0.0302.Vw2 - 0.622.DOC2.

The output obtained from ABC algorithm is consistent (Vw = 5m/min, and DOC = 6 μm), for all three replications (runs). The convergence of the ABC algorithm is shown in Fig. 6. 80

Ra = 0.924 - 0.0489.Vw - 0.008. DOC - 0.00031.Vw.DOC + 0.00172.Vw2 +0.00182.DOC2.

Table 5. Control parameters for ABC simulation Control parameter Value Swarm size

10

Number of employed bees

50% of the swarm size

Number of onlookers

50% of the swarm size

Number of scouts per cycle

1

Number of cycles

2000

Number of runs

3

When the ABC algorithm was used for optimization, initial tuning of the algorithm was done. It is well known that by reducing the swarm size, computation time also reduces. However, it adversely affects the solution quality. After testing for different swarm sizes during tuning, we finalized on swarm size of 10 as an acceptable trade-off between solution time, and quality. We also set the ratio of onlooker bees to 50% as per practice reported by other researchers [28], [31]; who suggested that this ratio provides the ideal trade-off between exploration and exploitation to obtain near-optimal solutions. Number of cycles was kept to 2000 to allow for sufficient iterations for the search to converge. Number of runs was set to three to ensure the repeatability of the search.

error Error

Coefficient of determination (R2(adj.)) values for this model were 91.48, 92.43 and 83.39 for Ft, Fn and Ra respectively. A reduced model were also generated, but not reported, as the R2 (adj.) was maximum for the complete regression model for Ft and Ra. However, for Fn, one of the reduced model exhibited the maximum R2 (adj.) in comparison with the complete model. This reduced model was a polynomial model, and its difference with the coefficients of the complete model was not that significant. Additionally, the reduced model and the complete model exhibited significant differences in coefficients from the linear model. The level of significance was taken at 5% for all comparisons. Results were then optimized using an ABC algorithm for a multi-objective scenario, where all the individual models were combined. Simulation of weighted-sum method of ABC was used in which the weights were taken as 1:1:1. The dependent variables were combined as the objective was to minimize the influence of independent variables on the response variable. The ABC algorithm was coded using MATLAB 6.5, and the values of the control parameters were chosen for easy convergence of the optimization problem, as given in Table 5.

75

70

65

0

5

10

15

20

25

cycles Cycles

30

35

40

45

50

Fig. 6. Convergence of ABC algorithm.

In this study, the values of the independent variables that minimizes the forces (energy requirements) and surface roughness (increase the quality) identified by ABC happens to be integer values. However, when individual models are considered, the optimal values could be non-integer values, like in the case of Ra, where the values are Vw = 11.56 m/min, and DOC = 6 μm. Early convergence of ABC can be attributed to the well behaved experimental data. Since, this is an initial step of the future work; where different unit processes will be combined to form a complete manufacturing process, whose greenness can be optimized using the approach demonstrated here. The authors hypothesize that for such large and complicated systems; multi-objective optimizing techniques like non-dominant curve method in ABC algorithm may be applied. This research demonstrated that the MQL method decreases the machining cost through significant reduction in cutting fluid costs, consumes lesser energy, and also reduces ecological hazards caused from the disposal of nonbiodegradable cutting fluids. This approach can become one of the many techniques for improving Green Index of a unit process, and in future of that of a complete manufacturing system. 4. Summary In this paper, a unit manufacturing process was studied by doing live experiments, and the experimental data was statistically analyzed to identify the significant factors influencing the green index of a process. Simulation was used to determine the optimum settings of the process parameters. This approach can be adapted for quantifying the greenness of a manufacturing facility. At this juncture, it cannot be claimed that the results can be directly transferred to other processes; however, the approach is transferable. Other optimization techniques like Particle Swarm Optimization (PSO), Bacterial Colony Algorithm (BCA), Genetic Algorithm (GA) etc. may also be compared with ABC to determine the efficacy of the approach. The authors are working on the sensitivity analysis, which at this point is missing from this study.

262

Amandeep Singh et al. / Procedia CIRP 29 (2015) 257 – 262

References [1] Glavic Peter, Lukman Rebeka. Review of sustainability terms and their definitions. Journal of Cleaner Production 2007; 15:1875-188. [2] Deif Ahmed M. A system model for green manufacturing. Journal of Cleaner Production 2011;19:1553-1559. [3] Duflou Joost R, Sutherland John W, Dornfeld David, Herrmann Christoph, Jeswiet Jack, Kara Sami, Hauschild Michael, Kellens Karel. Towards energy and resource efficient manufacturing:A processes and systems approach. CIRP Annals Manufacturing Technology.2012;61:587-609. [4] Dangelico Rosa Maria, Pontrandolfo Pierpaolo. From green product definitions and classifications to the Green Option Matrix. Journal of Cleaner Production. 2010;18:1608-1628. [5] Thiede S, Posselt G, Herrmann C. SME appropriate concept for continuously improving the energy and resource efficiency in manufacturing companies. CIRP Journal of Manufacturing Science and Technology. 2013;6:3:204-211. [6] Ghadimi Pouya, Li Wen, Kara Sami, Herrmanna Christoph. Integrated Material and Energy Flow Analysis towards Energy Efficient Manufacturing. 21st CIRP Conference on Life Cycle Engineering. Procedia CIRP. 2014;15:117-122. [7] Kellens K, Dewulf W, Overcash M, Hauschild M, Duflou J R. Methodology for Systematic Analysis and Improvement of Manufacturing Unit Process Life Cycle Inventory. Part 1:Methodology Description. International Journal of Life Cycle Assessment. 2012; 17:1:69-78. [8] Borgia Stefano, Pellegrinelli Stefania, Bianchi Giacomo, Leonesio Marco. A reduced model for energy consumption analysis in milling Variety Management in Manufacturing. Proceedings of the 47th CIRP Conference on Manufacturing Systems. Procedia CIRP. 2014; 17:529534. [9] Aurich J C, Linke B, Hauschild M, Carrella M, Kirsch B. Sustainability of abrasive processes. CIRP Annals - Manufacturing Technology. 2013; 62:653-672. [10] Hashimoto F. Specific Energy Comparison between Hard Turning and Grinding, STC G Meeting, Paris.2010. [11] Xiao K Q, Zhang L C. The effect of compressed cold air and vegetable oil on the subsurface residual stress of ground tool steel. Journal of Materials Processing Technology.2006;178:9-13. [12] Gutowski T G, Murphy C, Allen D, Bauer D, Bras B, Piwonka T, Sheng P, Sutherland J, Thurston D, Wolff E. Environmentally benign manufacturing: observations from Japan, Europe and the United States. Journal of Cleaner Production. 2005;13:1-17. [13] Winter Marius, Li Wen, Kara Sami, Herrmann Christoph. Determining optimal process parameters to increase the eco-efficiency of grinding processes, Journal of Cleaner Production. 2014;66:644-654. [14] Winter Marius, Bock Ralf, Herrmann Christoph. Investigation of a new ecologically benign metalworking fluid in abrasive machining processes to substitute mineral oil based fluids, 5th CIRP Conference on High Performance Cutting, Procedia CIRP.2012;1:393-398. [15] Winter Marius, Herrmann Christoph. Eco-Efficiency of alternative and conventional cutting fluids in external cylindrical grinding, Procedia CIRP. 2014;15:68-73. [16] Li Wen, Winter Marius, Kara Sami, Herrmann Christoph. Ecoefficiency of manufacturing processes: A grinding case, CIRP Annals Manufacturing Technology. 2012;61:59-62. [17] Obikawa Toshiyuki, Kamata Yasuhiro, Shinozuka Jun. High-speed grooving with applying MQL. International Journal of Machine Tools & Manufacture.2006; 46:1854-1861. [18] Sadeghi M H, Haddad M J, Tawakoli T, Emami M. Minimal quantity lubrication-MQL in grinding of Ti6Al4V titanium alloy, International Journal of Advance Manufacturing and Technology. 2009; 44:487-500. [19] Fogel L J, Owens A J, Walsh M J. Artificial intellgence through a simulation of evolution. Biophysics and Cybernetic systems. Proceeding of the 2nd Cybernetic Sciences Symposium .1965;131-155. [20] Goldberg D. Genetic Algorithms in Search Optimization and Machine Learning, Addison Wesley Publishing Company. Reading. Massachutes.1986. [21] Kennedy J, Eberhart R C. Particle Swarm Optimization, Proceeding of IEEE International Conference on Neural Networks.Perth, Australia, IEEE Service Center. Piscataway. NJ.1995;1942-1948.

[22] Price K, Storn R. Differential Evolution - a Simple and Efficient Adaptive Scheme for Global Optimization Over Continuous Spaces, Technical Report, International Computer Science Institute, Berkley. 1995. [23] Gambardella Luca M, Dorigo Marco. Ant-Q: A Reinforcement Learning approach to the traveling salesman problem Machine Learning Proceedings 1995. Proceedings of the Twelfth International Conference on Machine Learning. Tahoe City. California. 1995; 252-260. [24] Kim Dong Hwa, Abraham Ajith, Cho Jae Hoon. A hybrid genetic algorithm and bacterial foraging approach for global optimization. Information Sciences. 2007; 177:18:3918-3937. [25] Kirkpatrick S, Gellat C D, Vecchi M P. Optimization by simulation annealing. Science New series. 1983; 220:4598: 671-680. [26] Karaboga D, Basturk B. A Powerful and Efficient Algorithm for Numerical Function Optimization: Artificial Bee Colony (ABC) algorithm. Journal of Global Optimization. Springer Netherlands. 2007; 39:459-471. [27] Karaboga D, Basturk B. Artificial Bee Colony (ABC) Optimization Algorithm for Solving Constrained Optimization Problems. LNCS: Advances in Soft Computing: Foundations of Fuzzy Logic and Soft Computing. Springer-Verlag. IFSA. 2007; 789-798. [28] Samanta S, Chakraborty S. Parametric optimization of some nontraditional machining processes using artificial bee colony algorithm Engineering Applications of Artificial Intelligence.2011; 24:946-957. [29] Rao R V, Pawar P J. Grinding process parameter optimization using non-traditional optimization algorithms. Proceedings of the Institution of Mechanical Engineers, Journal of Engineering Manufacture. 2010;224:887-898. [30] Mukherjee Rajarshi, Chakraborty Shankar, Samanta Suman. Selection of wire electrical discharge machining process parameters using nontraditional optimization algorithms, Applied Soft Computing. 2012;12:2506-2516. [31] Yildiz Ali R. A new hybrid artificial bee colony algorithm for robust optimal design and manufacturing, Applied Soft Computing. 2013; 13:2906-2912 [32] Yildiz Ali R. Optimization of cutting parameters in multi-pass turning using artificial bee colony-based approach, Information Sciences. 2013; 20:399-407. [33] Yusup Norfadzlan, Zain Azlan Mohd, Hashim Siti Zaiton Mohd. Evolutionary techniques in optimizing machining parameters: Review and recent applications (2007–2011), Expert Systems with Applications. 2012;39:9909-9927. [34] Rao R Venkata, Pawar P J. Parameter optimization of a multi-pass milling process using non-traditional optimization algorithms, Applied Soft Computing. 2010;10:445-456.