Design of Experiments in Campaign Logistics Analysis

PurposeThe purpose of this paper is to focus the application of design of experiments (DOE) using industrial equipments, reinforcing idea that non‐statistical aspects in planning and conducting experiments are so important as formal design and analysis.Design/methodology/approachTwo case studies are presented to illustrate typical industrial applications and difficulties. Supported on these case studies and literature, this paper presents guidelines to planning, conducting and analysis involving technical and organizational aspects.FindingsSolving problems in industry, including in companies recognized as competent in the respective industrial sector, is not just a question of applying the right technique. Ceramic industry case study illustrates how important are non‐statistical issues in DOE application. Paint industry case study illustrates the strong relationship of the results with incorporating presented guidelines into practice. Moreover, both case studies consolidating a fundamental advantage of DOE: experimentation provides more knowledge about products, processes and technologies, even in unsuccessful case studies.Research limitations/implicationsUnsuccessful cases studies are very useful for identifying pitfalls and others limitations. This paper highlight difficulties aroused from non‐statistical aspects, although it is possible to find unsuccessful case studies due to statistical issues also. So, papers illustrating inadequate application of statistical techniques are welcome.Practical implicationsSuccessful DOE implementation depends on statistical and non‐statistical aspects. Although none of them shall be neglected, technical skills and technological knowledge about processes and products, management understanding of potential possibilities of statistical techniques and statistical fundamentals and knowledge about techniques of DOE must be ensuring to successful case studies in industrial setting.Originality/valueThis paper highlights non‐statistical aspects instead of the statistical ones. Tob overcome difficulties structured guidelines were designed to support DOE application in industrial setting.

Resistance spot welding (RSW) is one of the most commonly used form of manufacturing process. To efficiently search for the optimal RSW parameters in a new welding power control scheme to obtain the maximum tensile strength of welded joints, a design of experiments (DoE) using orthogonal arrays has been conducted. However the lack of success of conventional DoE is attributed to two facts that the weld quality is quantitatively measured and level-based evaluated, and the welding expulsion compromises the continuity of response surface, which is an underlying hypothesis of normal DoE analysis. Hereby a fuzzy analysis methodology for the experimentation design, referred as fuzzy orthogonal experiments (FOE) in this paper, has been proposed. With the fuzzification of factor levels and experimental objectives, the proposed method is able to explore the results of arbitrarily given factors without additional experiments, and identifies the varying tendency under condition of discontinuity experimental responses. The FOE application for a specifically planned RSW process presents a detailed FOE analyzing procedures, and yields the optimal parameter settings. The case study shows that FOE outperforms the traditional orthogonal DoE for its minimized experiment numbers and robust response tendency analysis.

In this paper, we develop an approach that determines the overall best parameter setting in design of experiments. The approach starts with successive orthogonal array experiments and ends with a full factorial experiment. The setup for the next orthogonal-array experiment is obtained from the previous one by either fixing a factor at a given level or by reducing the number of levels considered for all currently non-fixed factors. We illustrate this method using an industrial problem with seven parameters, each with three levels. In previous work, the full factorial of 37 = 2,187 points was evaluated and the best point was found. With the new method, we found the same point using 3% of these evaluations. As a further comparison, we obtained the optimum using a traditional Taguchi approach, and found it corresponded to the 366th of the 2,187 possibilities when sorted by the objective function. We conclude the proposed approach would provide an accurate, fast, and economic tool for optimization using design of experiments.

Optimizing parameters of freeform surface reconstruction using CMM

Recently, problems with excessive vibration of the radar masts of large bulk carriers and crude oil tankers have frequently been reported. This paper explores a design method to avoid the resonance of a radar mast installed on a large ship using various design of experiment (DOE) methods. A local vibration test was performed during an actual sea trial to determine the excitation sources of the vibration related to the resonant frequency of the radar mast. DOE methods such as the orthogonal array (OA) and Latin hypercube design (LHD) methods were used to analyze the Pareto effects on the radar mast vibration. In these DOE methods, the main vibration performances such as the natural frequency and weight of the radar mast were set as responses, while the shape and thickness of the main structural members of the radar mast were set as design factors. From the DOE-based Pareto effect results, we selected the significant structural members with the greatest influence on the vibration characteristics of the radar mast. Full factorial design (FFD) was applied to verify the Pareto effect results of the OA and LHD methods. The design of the main structural members of the radar mast to avoid resonance was reviewed, and a normal mode analysis was performed for each design using the finite element method. Based on the results of this normal mode analysis, we selected a design case that could avoid the resonance from the major excitation sources. In addition, a modal test was performed on the determined design to verify the normal mode analysis results.

An efficient design method for flip chip ball grid array (BGA) packages has been developed. This method uses design of experiment (DOE), a series of finite element (FE) stress analyses based on an orthogonal array used in DOE, and statistical analysis. By using this method to design a BGA package having 1600 pins, the warpage of the optimum packaging structure is very small and all of the reliability demands are satisfied.

A new intracellular antifungal protein (afp) production with average molecular weight 24.3 kDa and yield of 0.65 ± 0.1 mg/gram dry cell weight (gdcw) of mycelia in submerged fermentation of Aspergillus giganteus MTCC 8408 was optimized. Taguchi's DOE (design of experiment) L27 orthogonal array (OA) was constructed using Qualitek-4 software with 8 most influensive factors namely, culture pH, temperature, slant age, inoculum volume, agitation and KH2PO4. Scanning electron microscopy (SEM) was used to correlate the effect of selected factors on fungal cell morphology and afp production. The crude protein purification was accomplished using pure ammonium sulfate fractionation followed by carboxymethyl cellulose (CMC) ion-exchange chromatography and sephadex G-100 gel filtration. The average molecular mass of the purified protein was figured by silver stained SDS (sodium dodecylsulphate)-PAGE (poly-acryl amide gel electrophoresis). In vitro antifungal susceptibility assay was profiled against Candida albicans NCIM 3471 and minimum inhibitory concentrations (MICs) were in the range 3 to 4 µg/ml. Characterization of protein was observed with FTIR (Fourier transform infrared spectroscopy) analysis. The optimal production condition for crude afp was obtained as follows: soluble starch: 20 g/l; Corn steep liquor (CSL, 2%) + proteose peptone (PP, 1%): 30 g/l; pH: 5.8; temperature: 25°C; slant age: 3 d; inoculum size: 5% (v/v); agitation: 180 rpm; KH2PO4: 0.1 g/l. The validation experiments using optimized conditions confirmed an improvement in afp production by 59.4% against the expected enhancement of afp production by 61.22%. The present statistical optimization study revealed an opportunity to promote economical design at the industrial level for future scale up of effective antifungal agent against opportunistic oral and vaginal infection.

One of the basic components of Industry 4.0 is the design of a flexible manufacturing system (FMS), which involves the choice of parameters to optimize its performance. Discrete event simulation (DES) models allow the user to understand the operation of dynamic and stochastic system performance and to support FMS diagnostics and design. In combination with DES models, optimization methods are often used to search for the optimal designs, which, above all, involve more than one objective function to be optimized simultaneously. These methods are called the multi-objective simulation–optimization (MOSO) method. Numerous MOSO methods have been developed in the literature, which spawned many proposed MOSO methods classifications. However, the performance of these methods is not guaranteed because there is an absence of comparative studies. Moreover, previous classifications have been focused on general MOSO methods and rarely related to the specific area of manufacturing design. For this reason, a new conceptual classification of MOSO used in FMS design is proposed. After that, four MOSO methods are selected, according to this classification, and compared through a detailed case study related to the FMS design problem. All of these methods studied are based on Design of Experiments (DoE). Two of them are metamodel-based approaches that integrate Goal Programming (GP) and Desirability Function (DF), respectively. The other two methods are not metamodel-based approaches, which integrate Gray Relational Analysis (GRA) and the VIKOR method, respectively. The comparative results show that the GP and VIKOR methods can result in better optimization than DF and GRA methods. Thus, the use of the simulation metamodel cannot prove its superiority in all situations.

This study investigated the optimization of a CNC turning process for SKD11 (JIS). The design of experiments (DOE) method with an orthogonal array was applied. Nine experimental runs were performed based on the orthogonal array. The surface properties of roughness average and roughness maximum and the roundness were studied; the analysis of variance (ANOVA) was adapted to investigate which parameters had the most influence on the CNC turning process for SKD11 (JIS). Also, models for predictions of the roughness average, roughness maximum, and roundness had been developed in terms of cutting speed, feed rate, depth of cut, and the mixture ratio of cutting fluids.

The objective of this work was to increase laccase production by Pleurotus ostreatus PLAB through culture medium optimization using solid state culture conditions. Increased laccase activity was obtained through design of experiments (DOE) using the Taguchi orthogonal array (OA). Seven factors, viz. lignocellulose, glucose, yeast extract, peptone, KH(2)PO(4), MgSO(4) 7H(2)O and MnSO(4) H(2)O at three levels and pH at two levels. OA layout of L18 (2(1) x 3(7)) was selected for the proposed experimental design using Minitab 17 software. Data analysis showed that lignocellulose (20 %) and glucose (10 g L1) had positive effect, whereas KH(2)PO(4), MgSO(4)∙7H(2)O and MnSO(4)∙H(2)O did not have significant effect on laccase production. Taguchi OA analysis showed that pH 6, lignocellulose 20 %, glucose 10 g L(-1), yeast extract 6 g L(-1), peptone 15 g L(-1), KH(2)PO(4) 3 g L1, MgSO(4)∙7H(2)O 0.5 g L(-1) and MnSO(4)∙H(2)O 0.1 g L-1 were the optimal conditions to maximize laccase production. The model predicted a 30.37 U g(-1) dry wt., which agreed with the experimentally obtained laccase activity 29.15 U g(-1) dry wt. at optimal conditions.

The use of high dielectric permittivity in vertical double gate MOSFET (VDG-MOSFET) can overcome the problem of power dissipation and leakage current. In this paper, we investigated the performance potential of Titanium dioxide (TiO2) which has been deposited on silicon for use as a high-permittivity gate insulator and tungsten silicide (WSi2) for use as metal gate based on 2-D numerical simulations. The analysis and optimization of n-channel VDG-MOSFET is conducted, using a combination of L9 orthogonal array (OA) of Taguchi method and grey relational analysis (GRA). Four process parameters which are VTH implant energy, halo implant dose, source/drain (S/D) implant dose and S/D implant tilt angle are optimized to obtain the desired value of on-current (ION), off-current (IOFF) and subthreshold slope (SS). The design of experiment (DoE) is based on the L9 OA of Taguchi method and then the experimental data for multiple responses are converted into a single unit called grey relational grade (GRG). The most optimal level of four process parameters towards multiple responses are selected based on the highest GRG. Based on the analysis of variance (ANOVA), the most significant factor is observed to be S/D implant dose with 35.6% factor effect on GRG. The optimized value for ION, IOFF and SS after the optimization approach are 1599.3 µA/µm, 865.5 pA/µm and 68.02 mV/dec respectively with 0.8362 of GRG.

Earth observation satellite system (EOSS) is the main space platform collecting ground information. Optimization of EOSS is difficult, as it is a complex system referring a great deal of design variables and uncertain factors. Therefore, an optimization framework based on design of experiment and surrogate model is proposed. Design of experiment is used to generate simulation plan, which will greatly cut down cost of simulation. Then surrogate model is built to analyze simulation data and approximate real EOSS. Genetic algorithm and improved general pattern search method are adopted to solve the model. According to the framework, a case study is carried out. The final results illustrate the framework is useful and effective for the problem of EOSS optimization.

As a result of a successful case study, we advance a methodology for combining design of experiments (DOE) and cumulative score (Cuscore) charts to control and monitor an industrial process subject to a feedback control scheme. We use DOE to screen out the most important variable from a larger set of variables. We use the critical variable found in the DOE phase as a compensating factor in a feedback controller. We derive the proper Cuscore statistic that will identify spike signals in a dynamic system that is subject to feedback control. We then apply this Cuscore statistic to the output error from the control scheme. We illustrate the effectiveness of this process improvement methodology using an industrial case study where we identify and eliminate a recurring output quality problem in a pleating and gluing manufacturing operation.

Design of Experiments (DoE) is a powerful technique for process optimization that has been widely deployed in almost all types of manufacturing processes and is used extensively in product and process design and development. There have not been as many efforts to apply powerful quality improvement techniques such as DoE to improve non-manufacturing processes. Factor levels often involve changing the way people work and so have to be handled carefully. It is even more important to get everyone working as a team. This paper explores the benefits and challenges in the application of DoE in non-manufacturing contexts. The viewpoints regarding the benefits and challenges of DoE in the non-manufacturing arena are gathered from a number of leading academics and practitioners in the field. The paper also makes an attempt to demystify the fact that DoE is not just applicable to manufacturing industries; rather it is equally applicable to non-manufacturing processes within manufacturing companies. The last part of the paper illustrates some case examples showing the power of the technique in non-manufacturing environments.

Multi-response optimization problems investigation is a crucial element in initiatives designed to enhance quality and overall productivity for manufacturing processes. Since no particular algorithm can be employed for all multi-response problems, defining the method that is utilized as a problem-solving technique is a vital step in the process factors optimization. Identifying a formal procedure of implementing the improvement approach in a multi-criteria decision-making problem is a critical need to ensure the consistency and sustainability of the enhancement methods. In this study, a Plan–Do–Check–Act (PDCA) framework is implemented for a case study in the food industry under which a multi-response optimization problem is investigated. The design of experiment (DOE) is used to examine the effect of process parameters on the quality responses by using the Taguchi method to find the optimal setting for each parameter. An orthogonal array (OA) and signal-to-noise (SNR) ratio is employed to investigate the performance characteristics. Each performance characteristic is then converted into a signal-to-noise ratio, and all the ratios are then fed into a fuzzy model to produce a single comprehensive output measure (COM). The average COM values for various factor levels are calculated, and the level that maximizes the COM value for each factor is identified as the optimal level. Results indicated the effectiveness of the applied method to find the optimal factor levels for the multi-response optimization problem under study. The global optimal factor levels that are driven from the fuzzy logic for the studied parameters are 1250, 40, 7.5, and 1:2, for the speed, frying time, cooking time, and the coating ratio, respectively. Means of all the studied quality characteristics were closer to the target values when compared with the initial factors’ settings.