with Applications in the Theory of Attractive Quality and Six Sigma

(1)

HABILITATION THESIS

Contributions on Development of Quality Engineering and Management

with Applications in the Theory of Attractive Quality and Six Sigma

Methodology

Research field: Engineering and Management Adrian Pavel Pugna, Ph.D.

2020

(2)

2

List of Figures

Fig. 1.1.1 One-dimensional Quality………... 22

Fig. 1.1.2 Two-dimensional Quality………... 22

Fig. 1.1.3 Quality types according to the Kano requirements categories…... 24

Fig. 1.1.4 Kano requirements categories over time…... 24

Fig. 1.1.5 Tabulation of Surveys…... 27

Fig. 1.1.6 Functional vs. Dysfunctional and Kano transformation………... 33

Fig. 1.1.7 Plots of Average Functionality and Average Dysfunctionality Points for Question J……... 34

Fig. 1.1.8 Plots of Average Functionality and Average Dysfunctionality Points for Question J with Average Importance Indicated for Each Point……... 35

Fig. 1.1.9 Hierarchy questionnaire (example)…... 35

Fig. 1.1.10 Importance questionnaire and Hierarchy questionnaires (own and competitor’s)……... 36

Fig. 1.2.1 The HWWP’s domains and Maslow’s hierarchy of needs……... 41

Fig. 1.2.2 The HWWP, a refined IVA-Kano model…... 42

Fig. 1.2.3 The HWWP model for the new shoe customization service…... 50

Fig.1.3.1 The HWWP model (Potra, Izvercian, Pugna, & Dahlgaard, 2015)…. 54 Fig. 1.3.2 The representation of the customer stated importance and value- added variables as a non-uniform model with elasticity curves…….. 58

Fig. 1.3.3 The quality attributes of the contest computed in an HWWP model with normalized importance of wants (X)…... 60

Fig. 1.3.4 The allocation of quality attributes to a specific elasticity curve in the refined HWWM model……... 62

Fig. 1.4.1 The Greenhouse Model…... 70

Fig. 1.4.2 The Greenhouse Model with respondents’ perception of importance and Berger’s satisfaction coefficient as the Value……. 72

Fig. 1.4.3 The square-shaped Greenhouse Model with the averages (Xavg, Yavg) positioned……... 76

Fig. 1.4.4 Attribute 8 (customizing accessories)…... 80

Fig. 1.4.5 Attribute 9 (view of other customer orders)…... 80

Fig. 1.4.6 Attribute 11 (customizing color)…... 81

Fig. 1.4.7 Attribute 13 (payment options)…... 81

Fig. 1.5.1 The methodology proposed in the study…... Fig. 1.5.2 The non-uniform partition of the HWWP model…... 91 96 Fig. 1.5.3 The HWWP model (co-creation contest)…... 100

Fig. 1.5.4 Records in polar coordinates according to Rayleigh’s z test…... 101

(5)

5

Fig. 1.5.5 The 14 quality attributes in the HWWP model…... 103

Fig. 1.6.1 The new integrated HWWP – DDDI model…... 111

Fig. 1.6.2 The 15 student’s requirements integrated into the HWWP-DDDI model…... 116

Fig. 2.1.1 Six Sigma and DMAIC Methodology overview (International Six Sigma Institute™)……... 118

Fig. 2.1.2 What is Six Sigma? (International Six Sigma Institute™)…. 119 Fig. 2.1.3 Six Sigma Process Excellence Disciplines(International Six Sigma Institute™)……... 120

Fig. 2.1.4 Six Sigma Standard Normal Distribution…... 121

Fig. 2.1.5 Six Sigma Standard Normal Distribution (short – term and long – term)…... 123

Fig. 2.1.6 Six Sigma Normal Distribution (shifted short – term and long – term)…... 123

Fig. 2.1.7 PDCA and DMAIC cycles…... 126

Fig. 2.1.8 Six Sigma DMAIC Process and Statistical Tools…... 126

Fig. 2.1.9 Six Sigma DMAIC Process and Management Tools…... 127

Fig. 2.1.10 Lean Six Sigma predecessors……... 128

Fig. 2.1.11 Lean Six Sigma and previous Operational Improvement eras (IBM Global Business Services analysis)……... 129

Fig. 2.2.1.1 Pareto Analysis for Upper Wire Horn Assembly…... 131

Fig. 2.2.1.2 Hand Operated Riveting Tool…... 131

Fig. 2.2.1.3 Assembled Rivet on Upper Wire Horn Assembly…... 132

Fig. 2.2.1.4 Process capability for Rivet Height…... 132

Fig. 2.2.1.5 R chart for Rivet Height…... 133

Fig. 2.2.1.6 Xbar chart for Rivet Height…... 134

Fig. 2.2.1.7 Ishikawa diagram for Rivet Height noncompliance…... 135

Fig. 2.2.1.8 FMEA for Rivet Height noncompliance…... 136

Fig. 2.2.1.9 Process capability for Rivet Height (improved process)…... 138

Fig. 2.2.1.10 R chart for Rivet Height (improved process)…... 138

Fig. 2.2.1.11 Xbar chart for Rivet Height (improved process)…... Fig. 2.2.1.12 Updated FMEA for Rivet Height noncompliance…... 138 139 Fig. 2.2.2.1 The SIPOC diagram for the WGA production system…... 143

Fig. 2.2.2.2 The flowchart for the WGA flow line…... 143

Fig. 2.2.2.3 The resulting element of station 2…... 144

Fig. 2.2.2.4 Pareto chart for the screwing process in station 2…... 145

Fig. 2.2.2.5 Range Chart fot Tightening Torque…... 147

(6)

6

Fig. 2.2.2.6 Xbar Chart for Tightening Torque…... 147

Fig. 2.2.2.7 Process Capability analysis for tightening torque…... 147

Fig. 2.2.2.8 The Ishikawa diagram for non-conforming assembly…... 149

Fig. 2.2.2.9 The automated production line flowchart…... 150

Fig. 2.2.2.10 R chart for the automated WGA production line…... 150

Fig. 2.2.2.11 R chart for the automated WGA production line…... 151

Fig. 2.2.2.12 Final process capability for the automated WGA production line.. 151

Fig. 2.2.3.1 Body Control Unit…... 158

Fig. 2.2.3.2 Microcontroller…... 159

Fig. 2.2.3.3 Data I/O programming simplified flowchart……... 159

Fig. 2.2.3.4 SMT/NXT programming simplified flowchart…... 159

Fig. 2.2.3.5 Pareto Analysis for Data I/O…... 161

Fig. 2.2.3.6 Initial capability analysis for the distance between pins (Gamma distribution)……... 163

Fig. 2.2.3.7 Initial capability analysis for the distance between pins (Normal distribution)……... 163

Fig. 2.2.3.8 Probability Plot for the distance between pins (Initial Capability Analysis using Gamma distribution)……... 164

Fig. 2.2.3.9 Probability Plot for the distance between pins (Initial Capability Analysis using Normal distribution)……... 164

Fig. 2.2.3.10 Xbar Chart for the distance between pins…... 166

Fig. 2.2.3.11 Range Chart for the distance between pins…... 166

Fig. 2.2.3.12 Pareto Analysis for SMT/NXT…... 168

Fig. 2.2.3.13 Ishikawa diagram for Data I/O…... 169

Fig. 2.2.3.14 Ishikawa diagram for SMT/NXT……… 170

Fig. 2.2.3.15 AHP hierarchy for selecting a roll supplier…... 172

Fig. 2.2.3.16 Priority Vector for criteria…... Fig. 2.2.3.17 Criteria weights…... 174 174 Fig. 2.2.3.18 Calculation of Eigenvalue λ^max…... 174

Fig. 2.2.3.19 Rolls with sprocket holes on one side (a) and both sides (b)…... 178

Fig. 2.2.3.20 Data I/O station optimization…... 178

Fig. 2.2.3.21 Final capability analysis for the distance between pins (Lognormal distribution)…... 180

Fig. 2.2.3.22 Final capability analysis for the distance between pins (Normal distribution)…... 181

Fig. 2.2.3.23 Xbar Chart for the distance between pins…... 182

Fig. 2.2.4.24 R Chart for the distance between pins…... 182

(7)

7

Fig. 2.2.3.25 Pareto Analysis for Data I/O after improvement…... 184

Fig. 2.2.3.26 Pareto Analysis for SMT/NXT after improvement…... 184

Fig. 3.1.1 Taguchi's strategy for minimizing the impact of noise factors……… 188

Fig. 3.1.2 Taguchi's approach to product design………... 189

Fig. 3.1.3 Using the Quality Loss Function………. 190

Fig. 3.1.4 Continuous Symmetric Quadratic Loss Function………. 191

Fig. 3.1.5 Quality Loss Function for a criterion to be minimized…... 192

Fig. 3.1.6 Quality Loss Function for a criterion to be maximized…... 193

Fig. 3.1.7 Linear graphs associated with the L⁸ matrix………. 197

Fig. 3.1.8 Taguchi combined experimental plan……… 198

Fig. 3.1.9 3^N full factorial experimental plan (27 points = 27 experiments)…… 199

Fig. 3.1.10 Three fractional factorial experiment plans 3^3-1(9 points = 9 experiments) from a complete factorial experiment plan 3³ (27 points = 27 experiments)……… 200

Fig. 3.1.11 CCD plan for 3 design variables at 2 levels……… 200

Fig. 3.1.12 Three-dimensional response surface and two-dimensional contour 202 Fig.3.1.1.1 General process of manufacturing sintered parts…... 204

Fig. 3.1.1.2 Basic phenomena occurring during sintering (Kang, 2005)…... 205

Fig. 3.1.1.3 EDAX Analyzer…... 209

Fig. 3.1.1.4 Components of basalt powder…... 210

Fig. 3.1.1.5 Basalt powder melting furnace…... 210

Fig. 3.1.1.6 Basalt powder inside the quartz capsule…... 211

Fig. 3.1.1.7 Parts of molten basalt…... 212

Fig. 3.1.1.8 The component elements of molten basalt……... 212

Fig. 3.1.1.9 (a), (b), (c), (d) Pressing mold…... 213

Fig. 3.1.1.10 (a), (b), (c), (d) Compacted parts obtained by pressing…... 214

Fig. 3.1.1.11 Drying/calcination oven…... 215

Fig. 3.1.1.12 Sintering / cooling furnace…... 215

Fig. 3.1.1.13 (a), (b), (c), (d) Sintered and cooled parts…... 216

Fig. 3.1.1.14 Components of sintered basalt…... 217

Fig. 3.1.1.15 (a), (b) Images of molten basalt…... 218

Fig. 3.1.1.16 (a), (b) Images of sintered basalt…... 219

Fig. 3.1.1.17 Controlled factors levels…... 219

Fig. 3.1.1.18 Experimental results and S/N ratios…... 220

(8)

8

Fig. 3.1.1.19 ANOVA table…... 220

Fig. 3.1.1.20 Significant factor and interaction influences…... 221

Fig. 3.1.1.21 Variation reduction plot-based an on assumed Normal Performance Distribution…... 224

Fig. 3.1.1.22 Variation reduction plots for the two confirmation experiments... 226

Fig. 3.1.1.23 Standardized Pareto chart for factor effects (excluding interactions)... 230

Fig. 3.1.1.24 Estimated Response Surface (Composition – Forming Pressure)... 236

Fig. 3.1.1.25 Estimated Response Surface Contours (Composition – Forming Pressure)... 236

Fig. 3.1.1.26 (a), (b) Device for supporting the sintered basalt pieces... 237

Fig. 3.1.1.27 (a), (b) Device for cutting sintered basalt pellets... 237

Fig. 3.1.1.28 (a), (b), (c), (d) Rectified braking disks... 238

Fig. 3.1.1.29 (a), (b) Drilling holes in braking disks... 238

Fig. 3.1.1.30 (a), (b) Inserting and fixing the basalt pellets in the brake disks.. 239

Fig. 3.3.1.31 Braking disk with 2 pads... 239

Fig. 3.3.1.32 Braking disk with 4 pads... 239

Fig. 3.3.1.33 Single disk mounting... 240

Fig. 3.3.1.34 Mounting with 2 pads... 240

Fig. 3.3.1.35 Mounting with 4 pads... 240

Fig. 3.1.1.36 Characteristic elements of the wheel and the braking disk... 241

Fig. 3.1.1.37 Characteristic elements of the pressure system... 242

Fig. 3.1.1.38 (a), (b), (c), (d) Practical realization of the pressure system... 244

Fig. 3.1.1.39 (a), (b) Wear measurement... 245

Fig. 3.1.1.40 Estimated (optimized) response surface for the average wear of the basalt pellets, abrasion disk D0... 248

Fig. 3.1.1.41 Estimated (optimized) response surface for the average wear of basalt pellets, abrasion disk D4... 248

Fig. 3.1.1.42 Estimated (optimized) response surface for the average wear of basalt pellets, abrasion disk D2... 249

Fig. 3.1.1.43 Histogram for the service life of sintered basalt pellets and the assigned distributions... 250

Fig. 3.1.1.44 Weibull first-order model assigned for the D0 abrasion disk (Revs maintained at 810 rpm)... 255

Fig. 3.1.1.47 Weibull first-order model assigned for the D0 abrasion disk (pressure maintained at 5.88 daN/cm²)... 256

(9)

9

Fig. 3.1.1.48 Weibull first-order model assigned for the D4 abrasion disk

(pressure maintained at 5.88 daN/cm²)... 257 Fig. 3.1.1.49 Weibull first-order model assigned for the D2 abrasion disk

(pressure maintained at 5.88 daN/cm²)... 257 Fig. 3.1.2.1 XRD spectra for P3M-H 200-Ag2-15 (a), P7M-H200-Ag3-

15 (b), P4M-H 200-Ag2-30 (c) and P8M-H200-Ag3-30 (d),

synthesized through M-H method... 260 Fig. 3.1.2.2 XRD spectra for P1M-H 150-Ag2-15 (a), P5M-H 150-Ag3-

15 (b), P2M-H 150-Ag2-30 (c) and P6M-H150-Ag3-30 (d),

synthesized through M-H method... 260 Fig. 3.1.2.3 SEM surface morphology (a); EDAX spectrum (b) for P4M-H (30- 200-1000)... 262 Fig. 3.1.2.4 Variation reduction plot for improved conditions based on normal distribution... 265 Fig. 3.1.2.5 Estimated Response Surface for Ag-doped TiO2 nanoparticles

dimensions – RSM (Draper-Lin) design... 268 Fig. 3.1.2.6 Standardized Pareto Diagram for factors effect……… 271 Fig. 3.1.2.7 Estimated Response Surface (dopant concentration - time

autoclaving)……….. 274

(10)

10

List of Tables

Tab. 1.1.1 Attribute categories according to the Kano Model………... 23

Tab. 1.1.2 Functional/dysfunctional questions and five possible answers in Kano survey ………... 25

Tab. 1.1.3 Evaluation of pair questions in Kano survey…... 25

Tab 1.1.4 Kano evaluation table…... 25

Tab. 1.1.5 Spreadsheet of Most Frequent Responses to Customer Requirements……... 28

Tab. 1.1.6 Spreadsheet of Customer Requirements Sorted in Order of Most Frequent Response………... 29

Tab. 1.1.7 Self-stated Importance Questionnaire……... 30

Tab 1.1.8 Modified Mode Statistic…... 30

Tab. 1.1.9 Example where Mode Statistic is inadequate……... 31

Tab. 1.1.10 Kano transformation table…... 32

Tab. 1.2.1 Table of results……... 49

Tab. 1.3.1 The 14 attributes of a personalization service with their coordinates, traditional Kano categories, HWWP dimensions, elasticity curve distances, and directions……... 61

Tab. 1.4.1 Respondents’ use of the Kano Questionnaire and the six alternative classifications………... 68

Tab. 1.4.2 Quality categories, satisfaction coefficient, importance, and greenhouse domains……... 72

Tab. 1.4.3 Suggested Value scale related to respondents’ use of the Kano Questionnaire…... 74

Tab. 1.4.4 Frequency distribution of 166 respondents’ perceptions related to new value scale…... 75

Tab. 1.4.5 Averages of importance and value with normalized scales {0;1}…. 76 Tab. 1.4.6 Model (1.4.4) – nonlinear case without intercept…... 79

Tab. 1.5.1 The 14 co-creation requirements with computed Kano categories.. 98

Tab. 1.5.2 The results for the contest requirements with the Kano category, SC, Importance variables, and the HWWP dimension…... 99

Tab. 1.5.3 Values for importance and added value variables using the Rayleigh test for directional statistics…... 101

Tab 1.5.4 Values for importance and added value variables using Chen and Hu test……... Tab. 1.5.5 The 14 attributes of the new contest service……... Tab. 1.6.1 The new application’s proposed quality attributes…... 102 113 Tab. 1.6.2 Results of the Kano questionnaire, SI, DI, importance questionnaire and ASC…... 115

Tab. 2.1.1 Connection between short-term Sigma Score Z^ST and the long- term DPMO……... 124

(11)

11

Tab. 2.1.2 Connection between C^P and C^PK……... 125

Tab. 2.1.3 Sigma Score and Yield, Out of Spec., C^PK, and DPMO (short-term).. 125

Tab. 2.1.4 Sigma Score and Yield, Out of Spec., C^PK = P^PK and DPMO (long- term)…... 125

Tab. 2.2.1.1 Statistical tests for normality………... 133

Tab. 2.2.1.2 AHP for selecting the most suitable supplier…... 137

Tab. 2.2.2.1 The project viability matrix……... 141

Tab. 2.2.2.2 Tests for normality for the tightening torque process…... 146

Tab. 2.2.2.3 Comparison of alternative distributions…... 146

Tab. 2.2.2.4 Five Whys analysis for the high number of scrap at the screwing station…... 148

Tab. 2.2.3.1 The viability matrix for the automotive process improvement project…... 157

Tab. 2.2.3.2 Data I/O non-conformities categories…... 160

Tab. 2.2.3.3 Tests for normality for the distance between pins…... 161

Tab. 2.2.3.4 Goodness-of-fit tests for the distance between pins…... 162

Tab. 2.2.3.5 Comparison of alternative distributions…... 163

Tab. 2.2.3.6 Percentages out of specifications for the distance between pins. 165 Tab. 2.2.3.7 Capability Indices for the distance between pins……… 165

Tab. 2.2.3.8 SMT/NXT non-conformities categories…... 167

Tab. 2.2.3.9 “5 Whys” analysis for adaptors and unprogrammed components.. 169

Tab. 2.2.3.10 “5 Whys” analysis for bent pins…... 170

Tab. 2.2.3.11 “5 Whys” analysis for SMT/NXT…... 171

Tab. 2.2.3.12 AHP for ranking scale for criteria and alternatives…... 172

Tab. 2.2.3.13 Criteria ranking for rolls supplier……….. 173

Tab. 2.2.3.14 Corresponding indices of consistency to the order of random matrix…... 175

Tab. 2.2.3.15 AHP for selecting the most suitable roll supplier…... 175

Tab. 2.2.3.16 Tests for normality for the distance between pins……... Tab. 2.2.3.17 Goodness-of-fit tests for the distance between pins…... 179 179 Tab. 2.2.3.18 Comparison of alternative distributions…... 180

Tab. 2.2.3.19 Percentages out of specifications for the distance between pins... 181 Tab. 2.2.3.20 Capability Indices for the distance between pins…... 181

Tab. 2.2.3.21 Data I/O non-conformities categories after improvement…... 183

Tab. 2.2.3.22 SMT/NXT non-conformities categories after improvement…... 183

(12)

12

Tab. 3.1.1 Standard Taguchi L⁸ matrix……….. 197

Tab. 3.1.1.1 Variables affecting sinterability and microstructure of sintered parts…... 205

Tab. 3.1.1.2 Oxide composition of basalt from Luncani, Romania, and optimum conditions... 207

Tab. 3.1.1.3 The order of importance of the factors…... 221

Tab. 3.1.1.4 Optimal levels…... 221

Tab. 3.1.1.5 Expected performance for the "Cooling duration" factor at level 2. 222 Tab. 3.1.1.6 Optimal expected values for S / N ratio and compression strength 223 Tab. 3.1.1.7 Current condition versus the improved condition – parameters comparison... 224

Tab. 3.1.1.8 Conditions for carrying out confirmation experiment number 1.... 225

Tab. 3.1.1.9 Conditions for carrying out confirmation experiment number 2.... 225

Tab. 3.1.1.10 Predicted conditions versus conditions obtained by confirmation experiments……….... 226

Tab. 3.1.1.11 The L⁸ orthogonal matrix obtained after simulating the complete factorial experiment... 227

Tab. 3.1.1.12 Optimal condition resulting from full factorial experiment simulation... 228

Tab. 3.1.1.13 Values from full factorial simulation and L⁸experiment – a comparison... 228

Tab. 3.1.1.14 Comparison between expected values based on the L⁸ orthogonal matrix and values obtained by simulation... 229

Tab. 3.1.1.15 Comparison between the order of importance of the factors for the Taguchi method and the RSM method (Draper-Lin)... 231

Tab. 3.1.1.16 Comparison between expected (predicted) values of the Taguchi and RSM methods (Draper-Lin plan)... 232

Tab. 3.1.1.17 Optimal factor levels of factors – comparison Taguchi L⁸ and Draper – Lin... 233

Tab. 3.1.1.18 Factor values and predicted value for compressive strength (change in composition in 0.1% increments)... 233

Tab. 3.1.1.19 Factor values and predicted value for compressive strength (change in forming pressure in increments of 25 daN/cm²)... 234

Tab. 3.1.1.20 The optimal combination of factor levels (with interactions) for the Draper-Lin plan... 234

Tab. 3.1.1.21 The optimal combination of factor levels (without interactions) for the Draper-Lin plan... 234

Tab. 3.1.1.22 Comparison between different methods of estimating, measuring, and calculating the value of compression strength... 235

Tab. 3.1.1.23 Data obtained from the wear tests... 245

Tab. 3.1.1.24 ANOVA for average wear... 246

Tab. 3.1.1.25 Sum of the third-order squares... 246

Tab. 3.1.1.26 Distributions that can model the service life of sintered basalt pellets... 249

(13)

13

Tab. 3.1.1.27 Tests to verify the normality of the data... 250

Tab. 3.1.1.28 Kolmogorov - Smirnov test for the adequacy of the assigned distributions... 250

Tab. 3.1.1.29 Cramer – Von Mises test for the adequacy of the assigned distributions... 251

Tab. 3.1.1.30 Watson test for the adequacy of the assigned distributions... 251

Tab. 3.1.1.31 Anderson – Darling test for ranking assigned distributions... 251

Tab. 3.1.1.32 Parameters estimation for the Weibull first-order regression model... 252

Tab. 3.1.1.33 Likelihood ratio tests for the factors of the Weibull first-order regression model... 252

Tab. 3.1.1.34 Parameters estimation for the Weibull second-order regression model... 253

Tab. 3.1.1.35 Likelihood ratio tests for the factors of the Weibull second- order regression model... 253

Tab. 3.1.1.36 Parameters estimation for the Weibull second-order regression model (modified)... 254

Tab. 3.1.1.37 Likelihood ratio tests for the factors of the Weibull second order regression model (modified)... 254

Tab. 3.1.2.1 Parameters of Ag-doped TiO2 probe synthesized through M-H method... 259

Tab. 3.1.2.2 Average dimensions of Ag-doped TiO2 nanoparticles synthesized through M-H method... 261

Tab. 3.1.2.3 Control factors and levels... 263

Tab. 3.1.2.4 Optimal condition based on the average effects of factors... 264

Tab. 3.1.2.5 Factors importance order... 264

Tab. 3.1.2.6 Parameters values at improved versus current conditions... 266

Tab. 3.1.2.7 Values comparisons between full factorial experiment simulation and L⁸... 267

Tab. 3.1.2.8 Factors importance order obtained through Taguchi and RSM designs... 268

Tab. 3.1.2.9 Parameters of Ag-doped TiO² samples, synthesized through F-H method……… 270

Tab. 3.1.2.10 Average dimensions of Ag-doped TiO² nanoparticles synthesized through F-H method………. 270

Tab. 3.1.2.11 Factor levels and experimental results under standard Taguchi L⁹ matrix……….. 271

Tab. 3.1.2.12 Measured values, predicted values, and their limits associated to L⁹……….. 272

Tab. 3.1.2.13 Optimized combination of factors levels for L⁹……….. 272

Tab. 3.1.2.14 Order of factors importance obtained through L⁹(Taguchi) and Box – Behnken (RSM) experimental plans designs……… 273

(14)

14

A. REZUMAT

Conceptul de cadru didactic universitar implicǎ atât o activitate didactică susținută cât și o activitate ştiinţificǎ validatǎ prin cărți, studii, cercetări și lucrări ştiinţifice.

Domeniul Ingineriei si Managementului Calității reprezintă un atribut esențial în dezvoltarea culturii inginerești și manageriale a viitorului inginer licențiat precum și a masterandului în inginerie și management.

Prezenta teză de abilitare relevă capacitățile și performanțele didactice și de cercetare ale candidatului dr.ing. Adrian Pavel Pugna, care ocupă în prezent o funcție didactică de conferențiar universitar în statul de funcțiuni al Departamentului de Management al Facultății de Manangement în Producție și Transporturi de la Universitatea Politehnica Timișoara, dezvoltate după susținerea publică a tezei de doctorat și până în prezent. Teza de abilitare se focalizeză, în principal, pe acele realizări care atestă capacitatea autorului de a desfășura și conduce activități de cercetare științifică în domeniul Inginerie și Management, cu aplicații în Teoria Calității Atractive și Metodologia Six Sigma.

În cei aproape 30 de ani de activitate didactică, candidatul și-a dezvoltat capacitățile și performanțele didactice, desfășurând toate tipurile de activități: seminar, laborator, proiect, curs, îndrumare la elaborarea de lucrări de diplomă și disertație etc., urcând scara ierarhică a funcțiilor didactice. Canditatul a contribuit la elaborarea de lucrări didactice, manuale universitare și îndrumătoare pentru lucrări aplicative.

În ceea ce privește activitatea managerială, candidatul a fost 5 ani membru în Consiliul Departamentului de Management, fiind membru al Biroului Consiliului Departamentului precum și coordonatorul comisiei pentru evaluarea și asigurarea calității și de asemenea, candidatul a fost 12 ani membru al Consiliului Profesoral al FMPT, unde a îndeplinit funcția de coordonator al comisiei pentru evaluarea și asigurarea calității. De asemenea, candidatul a fost (și este și în prezent) președintele Board-ului masterului Ingineria și Managementul Competitivității precum cum și membru în board-urilor masterelor Ingineria și Managementul Calității și Competitivității (limba engleză), Ingineria și Managementul Sistemelor Logistice, Management Antreprenorial în Administrarea Afacerilor.

Prezentarea rezultatelor obținute în activitatea de cercetare a candidatului ocupă cea mai mare parte a conținutului tezei de abilitare. Teza de abilitare este structurată

(15)

15

pe 3 direcții de cercetare importante și de mare actualitate în domeniul Ingineriei și Managementului, așa cum rezultă din literatura de specialitate:

1. Teoria Calității Atractive;

2. Metodologia Six Sigma;

3. Aplicații moderne ale proiectării experimentelor.

În prima parte a tezei se prezintă bazele Teoriei Calității Atractive precum și contribuțiile candidatului la dezvoltarea de noi modele principiale și aplicative în acest domeniu. În 2015, candidatul a contribuit la realizarea unui nou model pentru proiectarea produselor și serviciilor noi . Modelul HWWP (Health – Weapon - Wealth – Prospect) face legătura între piramida lui Maslow (Maslow’s hierarchy of needs), Modelul și metodologia Kano, importanța dorințelor clienților și coeficientul satisfacșiei clientului (SC). Acest model reprezintă o contribuție teoretică fundamentală la dezvoltarea modelului Kano și un punct de referință pentru cercetări ulterioare. În 2016 candidatul a contribuit la realizarea unui model HWWP rafinat, bazat pe partiția neuniformă cu curbe de elasticitate. În 2016, candidatul a contribuit la dezvoltarea unei abordări strategice pentru analiza variațiilor între nevoile potențiale ale clienților pentru o mai bună înțelegere ce ”elemente de calitate” trebuie cultivate înainte de lansarea produsului sau serviciului, numită, ” A Greenhouse Approach for Value Cultivation”

sau modelul Greenhouse. În 2020, candidatul a contribuit la realizarea unui model HWWP generalizat. Tot în 2020, candidatul a contribuit la realizarea unui nou model pentru evaluarea ”vocii studenților” în etapa de dezvoltare a unei aplicații pentru telefonul mobil, numit modelul HWWP – DDDI.

În partea a doua a tezei se prezintă elementele fundamentale ce stau la baza metodologiei Six Sigma. De asemenea se prezintă câteva din realizările candidatului în ceea ce privește aplicarea metodologiei Six Sigma în industria Automotive.

În partea a 3-a se prezintă elementele fundamentale ale proiectării experimentelor, cu accent pe metodologiile Taguchi și RSM (Response Surface Methodology). Sunt prezentate câteva din contribuțiile candidatului la utilizarea acestor metodologii la realizarea și încercarea pieselor din bazalt sinterizat precum și la realizarea nanopartipulelor de TiO² dopate cu argint.

(16)

16

Rezultatele activității de cercetare ale candidatului au fost prezentate în cadrul unor manifestări academice și științifice naționale și internaționale, prin articole publicate în reviste sau în volumele de lucrări ale conferințelor.

Candidatul a publicat un număr de 117 de articole. Distribuția pe categorii de publicaţii a acestor lucrări este după cum urmează:

• 4 în reviste indexate în baza de date Web of Science (Clarivate Analytics);

• 29 la conferințe internaționale indexate în baza de date Web of Science (Clarivate Analytics).

• 13 în reviste și volume indexate în alte baze de date internaționale (BDI);

• 71 în reviste sau volume neindexate în BDI;

De asemenea, candidatul este coautor la 1 carte la o editură internațională, coautor la 5 capitole în cărți la edituri internaționale, autor și coautor la 9 cărți la edituri naționale recunoscute și autor și coautor la 8 materiale didactice inclusiv în format electronic - suport de curs/îndrumare.

A doua parte a acestei secțiuni prezintă perspectivele de dezvoltare.

Ultima parte a acestei secțiuni prezintă referințele bibliografice.

(17)

17

A. ABSTRACT

The concept of University Professor involves both a sustained teaching activity and a scientific activity validated through books, studies, research, and scientific papers.

The field of Engineering and Quality Management is an essential attribute in the development of the engineering and managerial culture of the future licensed engineer as well as of the master student in engineering and management.

The present habilitation thesis reveals the didactic and research capacities and performances of the candidate Adrian Pavel Pugna, Ph.D. who currently holds a teaching position as an Associate Professor in the staff of the Department of Management of the Faculty of Management in Production and Transport at the University Politehnica Timisoara, developed after the public defense of his doctoral thesis.

The habilitation thesis focuses mainly on those achievements that attest to the author's ability to conduct and conduct scientific research in the field of Engineering and Management, with applications in Attractive Quality Theory and Six Sigma Methodology.

In the almost 30 years of teaching activity, the candidate has developed his teaching skills and performances, carrying out all types of activities: seminar, laboratory, project, course, guidance in the elaboration of diploma and dissertation papers, etc.

teaching functions. The candidate contributed to the elaboration of didactic works, university textbooks, and tutorials for applied works.

Regarding the managerial activity, the candidate was 5 years member of the Board of the Management Department, is a member of the Bureau of the Department Council as well as the coordinator of the commission for evaluation and quality assurance and also the candidate was 12 years member of the Faculty Council of FMPT, where he served as coordinator of the commission for evaluation and quality assurance. Also, the candidate was (and is currently) the chairman of the Board of the master's degree in Engineering and Competitiveness Management as well as a member of the boards of the masters of Engineering and Management of Quality and Competitiveness (English), Logistics Systems Management and Engineering in Business Administration.

(18)

18

The presentation of the results obtained in the research activity of the candidate occupies most of the content of the habilitation thesis. The habilitation thesis is structured on 3 important and highly topical research directions in the field of Engineering and Management, as it results from the specialized literature:

1. Theory of Attractive Quality;

2. Six Sigma Methodology;

3. Modern applications of experimental design.

The first part of the thesis presents the bases of the Theory of Attractive Quality as well as the candidate's contributions to the development of new and applied models in this research field. In 2015, the candidate contributed to the development of a new model for the design of new products and services. The HWWP (Health - Weapon - Wealth - Prospect) model which connects Maslow’s pyramid (Maslow’s hierarchy of needs), the Kano model and methodology, the importance of customer wants and the customer satisfaction coefficient (SC). This model represents a fundamental theoretical contribution to the development of the Kano model and a reference point for further research. In 2016 the candidate contributed to the realization of a refined HWWP model, based on the non-uniform partition with elasticity curves. In 2016, the candidate contributed to the development of a strategic approach to analyzing variations in potential customer needs for a better understanding of what “quality elements” need to be cultivated before launching the product or service, called “A Greenhouse Approach for Value Cultivation” or the Greenhouse model. In 2020, the candidate contributed to the realization of a generalized HWWP model. Also in 2020, the candidate contributed to the development of a new model for evaluating the

"student voice" in the development stage of a mobile phone application, called the HWWP - DDDI model.

The second part of the thesis presents the fundamental elements underlying the Six Sigma methodology. It also presents some of the candidate's achievements in the application of the Six Sigma methodology in the Automotive industry.

Part 3 presents the fundamentals of experimental design, with emphasis on Taguchi and RSM (Response Surface Methodology) methodologies. Some of the candidate's contributions to the use of these methodologies in the production and testing of

(19)

19

sintered basalt parts as well as in the production of silver-doped TiO² nanoparticles are presented.

The results of the candidate's research activity were presented in national and international academic and scientific events, through articles published in journals or the volumes of conference papers.

The candidate published 117 articles. The distribution by publication categories of these works is as follows:

• 4 in journals indexed in the Web of Science database (Clarivate Analytics);

• 29 at international conferences indexed in the Web of Science database (Clarivate Analytics).

• 13 in journals and volumes indexed in other international databases (BDI);

• 71 in journals or volumes not indexed in BDI;

Also, the candidate is co-author of 1 book at an international publishing house, co- author of 5 chapters in books at international publishing houses, author and co-author of 9 books at recognized national publishing houses, and author and co-author of 8 teaching materials including in electronic format - course support /guidance.

The second part of this section presents development perspectives.

The last part of this section presents the bibliographic references.

(20)

20

B. RESEARCH RESULT

1. Theory of Attractive Quality

1.1 Introduction

Voice of Customer (VOC) is the statement made by the customer on a particular product or service reflecting its voice, expectations, preferences, or comments (International Six Sigma Institute™, http://www.sixsigma-institute.org).

Capturing Voice Of Customer (VOC) is very important because customers are the ones who buy, use or transform either products or services and are the ones who receive the process output, regardless if they are internal customers (the ones who are internal to the organization, e.g Management, Employee(s) or Any Functional Department) or external customers (are not a part of the organization but use product(s) or service(s) or have invested interest in the organization, e.g. Clients, End- Users or Shareholders).

It is very important to distinguish between customer’s “needs” and “wants”. “Needs”

are desires or expectations of a customer from a given product or service Unfortunately are expressed often vague and generally are “wants” from a product/service. It is important to separate “needs” and “wants” because the first ones are critical features of products or services and the second ones are expectations that are beyond “needs”. If “needs” are not met by the products or services, the customers will be highly disappointed, and is a high probability they will switch to the competition.

If “wants are” not met, the customers may only be highly displeased or dissatisfied but will remain loyal to the brand, products, or services. “Requirements” are attributes of the product or service which fulfill the “needs” of customers. From the customer’s perspective, “requirements” are a “must” and therefore, as explained above, the customers may buy the products or services if “requirements” are met, even if the

“wants” are met or not met. Voice of Customer (VOC) methodology can be used to capture the customer needs – both current (stated needs) and latent (unstated needs).

VOC methodology helps capture the needs of the customer through stated verbatim comments (customer voices). It helps translate verbatim comments (customer voices)

(21)

21

into customer needs to product/service output characteristics (customer requirements).

1.1.1 Kano Model

KANO Analysis is about prioritizing customer requirements once they are established. Noriaki Kano (Kano et al., 1984) is the one who developed the "Customer Satisfaction Model" or Attractive Quality Model" (known as the Kano Model) which by a simple hierarchical scheme distinguishes between the essential and differentiating attributes related to the customer quality concepts.

Thus, the Kano Model focuses on differentiating the attributes of the product and not how it was initially done on the needs of the customers. He also proposed a methodology for connecting customer responses to a special questionnaire. It is very important for organizations, especially those working on the development of new innovative products and services, to know the needs and requirements of their customers as quickly as possible.

The "Creating the Attractive Quality" approach, also known as the "Kano Model", came about as a result of questioning the traditional idea that by acting more intensively on a product or service, the customer will be even more satisfied. Noriaki Kano argued that the performance of a product or service is not equal in the eyes of customers, in the sense that the performance of certain categories of attributes of the products or services produces higher levels of satisfaction than others (International Six Sigma Institute™, http://www.sixsigma-institute.org).

Noriaki Kano improved the definition of quality by adding a new dimension to it, the previous definitions of quality up to that time were linear and one-dimensional (Figure 1.1.1). Noriaki Kano integrated the quality of a two-dimensional model, considering two dimensions: how a product or service behaves in terms of performance (X-axis), and the degree of user/customer satisfaction (Y-axis), as in figure 1.1.2.

(22)

22

Fig. 1.1.1 One-dimensional Quality

Fig. 1.1.2 Two-dimensional Quality

If customers are asked in the exploration phase only about their desires and the reasons for the purchase, the results are usually disappointing and the answers already are known. It is very important to analyze "customer problems" instead of analyzing "customer wishes". If customers are asked in the exploration phase only about their desires and the reasons for the purchase, the results are usually disappointing and the answers already are known. The expectations regarding the product, mentioned by the customers, are only the tip of the iceberg, being necessary to highlight the "hidden" needs and problems. A detailed analysis of the problems to be solved, the conditions, and the environment in which the product evolves can lead to information on further product developments (International Six Sigma Institute™, http://www.sixsigma-institute.org).

The elaboration of the Kano Model starts with a survey of customers based on questionnaires, they are asked about the attributes of the products and what perception they have so much that they have them sufficiently or insufficiently. A Kano

(23)

23

survey asks 2 questions for each product attribute, resulting in the categories in table 1.1.1.

Tab. 1.1.1 Attribute categories according to the Kano Model

.

Basic Requirements are “Must-be’s”. They are the most important needs of customers. They are required and expected to be there. These are the needs the customer assumes will be met. When these requirements are unfulfilled, the customer is greatly dissatisfied and when fulfilled, the customer is neutral (i.e., they do not produce additional satisfaction).

For Performance Requirements there is a direct positive correlation that exists between satisfaction levels and the degree of presence. The more performance requirement elements needs are met, the better it is for the product or service.

Indifferent elements are needs that result in neither satisfaction nor dissatisfaction whether they are present / met or not.

Reverse elements are needs that result in either dissatisfaction when they are fulfilled or satisfaction even when they are not fulfilled.

Delighter Requirements are “attractors” and their presence in a product/process is unexpected and fulfill the latent needs of a customer. They lead to great satisfaction if found present and the customer still is neutral (& not dissatisfied) when absent.

Skeptic (or Questionable) refers to the fact that there is some uncertainty in the customer's response. Noriaki Kano follows these categories with the phrase "quality features", meaning a high functional level of the requirements (Figure 1.1.3).

(24)

24

Fig. 1.1.3 Quality types according to the Kano requirements categories

It is accepted that the positions of the Kano requirements categories are changing over time (Figure 1.1.4).

Fig. 1.1.4 Kano requirements categories over time

Another possibility to distinguish between the types of requirements that influence customer satisfaction is to answer in 5 different ways each pair of questions for each element of the product or service.

(25)

25

➢ Construction of Kano questionnaires

Table 1.1.2 presents functional/dysfunctional questions and five possible answers in the Kano survey and tables 1.1.3 and 1.1.4 presents the evaluation of pair questions in the Kano survey.

Tab. 1.1.2 Functional/dysfunctional questions and five possible answers in Kano survey

Tab. 1.1.3 Evaluation of pair questions in Kano survey

Tab 1.1.4 Kano evaluation table

(26)

26

➢ Testing Kano - type questionnaires

When a Kano questionnaire is sent to multiple clients, it must be understood by all.

First of all, ask the members who participate in the development of the Kano questionnaire to answer questions. Each member should think of a customer trying to predict what he or she would answer and what questions they might not understand.

Then select people from within the organization and give them the questionnaire.

Select a large variety of staff (managers, engineers, marketing staff, etc.). If the internal test signals something confusing, customers will likely notice the same thing.

Review and retest the questionnaire by adding, if necessary, additional instructions.

➢ Administration of Kano - type questionnaires

Select the customers you want to interview, ensuring the representativeness of the sample. Decide how you want to send the questionnaires: telephone, fax, mail, e-mail, face-to-face, etc. The most used is the transmission by mail (if you opt for this mode, write a cover letter explaining the purpose of the survey and include supplemental instructions for clients). If you also use Importance Questionnaires with Kano Questionnaires, use the same sequence of questions in both to compare the two questionnaires more easily. Keep a log of the customers to whom questionnaires were sent and record the results as they arrive.

➢ Processing the results

Responses for each customer requirement in a Kano Questionnaire are tabulated according to figures 1.1.5.

(27)

27

Fig. 1.1.5 Tabulation of Surveys

The easiest way to choose a category is to use the code that appears most often in the answers for a certain requirement (Customer Requirements – C.R.) - that is, the statistical mode of the answers is used. If any requirement receives a substantial number of Q (questionable) scores, the question must probably be temporarily removed from the analysis until the confusion with it is resolved or the respondents' thinking can be clarified. If most of the respondents give an R (reverse) scores to one of the requirements, this indicates that the market thinking about the question is opposed to the thinking of those who created the questionnaire.

➢ Analyzing the results

Benefits obtained by analyzing the data of the Kano questionnaire:

✓ A better understanding of customer requirements;

(28)

28

✓ Prioritizing the requirements for development activities;

✓ Detecting the characteristics of market segments;

✓ Helping the design process.

Answers should only be viewed as a guide, they do not provide exact answers as to which element should be included in the product or which requirement should not be fully met. It is usually desired to do a more in-depth analysis than the simple statistical mode.

When two Kano codes are equal in the evaluation of a question, it is considered:

✓ Contacting customers for additional information,

✓ Research for differences in market segmentation,

✓ Selecting the classification that will have the greatest impact on the product, using the order from relation (1.1.1).

M > O > A > I

(1.1.1) Another way to study the data is to build a table with columns for the first, second and third most common responses (Table 1.1.5). Then the rows can be rearranged into groups according to the order according to relation (1.1.1) (Table 1.1.6).

Tab. 1.1.5 Spreadsheet of Most Frequent Responses to Customer Requirements

(29)

29

Tab. 1.1.6 Spreadsheet of Customer Requirements Sorted in Order of Most Frequent Response

If it is used in parallel with the Kano questionnaire a Self-stated Importance questionnaire, then the answers can be sorted based on it. If there are several requirements of customers whose rating was A, then the data from the Importance questionnaire can be used to sort in descending order of importance of these requirements.

As a general guide, it may be considered necessary:

✓ Meeting the M requirements,

✓ To be competitive with market leaders relative to O,

✓ To include for differentiation elements A.

➢ The Self-stated Importance Questionnaire

For each potential customer requirement included in the Kano questionnaire, a question is constructed in the Self-stated Importance Questionnaire in the general format:

"How important is it or would it be if: [requirement X]? (Table 1.1.7).

(30)

30 Tab. 1.1.7 Self-stated Importance Questionnaire

➢ Improvements in the Analysis of Results

If the Kano questionnaire asks about very general functions, such as whether a machine must have 3 or 4 wheels or software must have a GUI (graphical user interface), each will have a specific opinion. However, if the Kano questionnaire asks about very specific functions, such as whether a machine must have ceramic valves or software must be compatible with a specific printer, then most respondents may be indifferent (I). Extremely detailed questions can increase the "noise level" to the level where all requirements are considered indifferent (I).

One way to change the simple statistical mode is if (O + A + M) > (I + R + Q), then the grade is maximum (O, A, M), if not, the grade is maximum (I, R, Q) (Table 1.1.8).

Tab 1.1.8 Modified Mode Statistic

There are some cases where the answers are "spread" over several categories or when divided differently into two categories, the grade is the same. In this case, the Mode Statistic is inadequate (Table 1.1.9). The idea is to reduce the data to 2 numbers, a positive one that gives the relative value of fulfilling the customers' requirements

(31)

31

(customer satisfaction will increase by providing elements A and O) and a negative one, showing the relative cost of not meeting that requirement (customer satisfaction will decrease if elements O and M are not included).

Tab. 1.1.9 Example where Mode Statistic is inadequate

➢ Customer Satisfaction Index – CSI

The Customer Satisfaction Index (CSI) is a method (proposed by Berger et al. in 1993) to identify attribute classification according to the Kano model. It consists of the rate of customers who declare to be satisfied with the presence or sufficiency of attributes (SI - Satisfaction Index), as well as the rate of customers who declare to be dissatisfied with the lack or insufficiency of the attributes (DI - Dissatisfaction Index).

✓ If SI > 0.5 and DI < 0.5, the attribute is classified as A.

✓ If SI ≤ 0.5 and DI ≥ 0.5, it is classified as M.

✓ If SI > 0.5 and DI > 0.5, it is classified as O.

✓ If SI < 0.5 and DI < 0.5, it is classified as I (Neutral).

Coefficients of Customer Satisfaction (SC), Dissatisfaction (DC), and Total Satisfaction are presented in relation (1.1.2), (1.1.3), and (1.1.4).

SC =

^A+O

A+O+M+I (1.1.2)

DC =

^O+M

(A+O+M+I)∙(−1) (1.1.3)

(32)

32

TC =

^A−M

A+O+M+I (1.1.4) The (-) sign in front of the DC coefficient is to emphasize its negative influence on customer satisfaction if the quality level for this product is not reached. The positive coefficient SC takes values between “0” and “1”; the closer the value to “1”, the greater the influence on customer satisfaction. An SC coefficient approaching “0” means that there is very little influence. If DC approaches “-1”, the influence on customer dissatisfaction is very strong if the product attribute is not fulfilled. A value of approximately "0" means that that attribute does not cause dissatisfaction if it is not fulfilled.

✓ If TSC < 0, then the requirements are O,

✓ If TSC > 0.1, then the requirements are A,

✓ If TSC = 0, then the requirements are R.

➢ Kano transformation table

The Kano transformation table is presented in table 1.1.10.

Tab. 1.1.10 Kano transformation table

The analysis form described in this section assumes the existence of Q pairs of questions, j = 1, ..., Q and N respondents, i = 1, ..., N. It is also assumed that the Kano Questionnaire is used in parallel with an Importance Questionnaire. Thus, there are three scores for each investigated customer requirement - Functional, Dysfunctional, and Importance. The 3 scores are coded as follows:

• Functional: Y^ij = -2 (Dislike), -1 (Live with), 0 (Neutral), 2 (Must-be), 4 (Like)

(33)

33

• Dysfunctional: X^ij = -2 (Like), -1 (Must be), 0 (Neutral), 2 (Live with), 4 (Dislike)

• Importance: W^ij = 1 (Not at all Important), ..., 9 (Extremely Important).

Note that X and Y take the values -2, -1, 0, 2, 4 The logic for the asymmetric scale (starting from -2, rather than from -4) is that "Must be" and "One-dimensional" are stronger answers than "Reverse" or "Questionable". That is why the scale should give less importance to less powerful responses to diminish their influence on the average.

Reverse type answers are given less importance, being "drawn" to zero. The purest representations for the Reverse, Indifferent, One-dimensional, Must-be, and Attractive points are identified in this coordinate system with the points:

• Reverse: X = - 2, Y = - 2

• Indifferent: X = 0, Y = 0

• One-dimensional: X = 4, Y = 4

• Must-be: X = 4, Y = 0

• Attractive: X = 0, Y = 4

These points are shown in figure 1.1.6 (underlined and bold). All other combinations of XY points appear as interpolations between these points.

Fig. 1.1.6 Functional vs. Dysfunctional and Kano transformation

(34)

34

Calculate for all questions, j = 1, ..., Q, the mean of X (dysfunctional) and Y (functional) as in relations (1.1.5) and (1.1.6).

X

_avg

[j] =

^{∑ X}ⁱ ^ij

N (1.1.5)

Y

_avg

[j] =

^{∑ Y}ⁱ ^ij

N (1.1.6) Represent the points Q (X^avg [j], Y^avg [j]), and use the number j as a symbol on the graph so that you can identify which question represents each point. Averages should fall between 0 and 4 because the negative values are either "Questionables" or

"Reverses". "Questionables" will not be included in the media. "Reverses" can be transformed from this category by changing the meaning of the functional and dysfunctional questions for all the respondents. Otherwise, as described above, there will not be enough "Reverses" to "pull" the media to the negative.

In figure 1.1.7, the square in which Xave and Yave are between 0 and 4 is divided into quadrants, considering the prototype points Attractive, One-dimensional, Must- be, and Indifferent placed in the 4 corners. This square comes from the upper right corner of the figure 1.1.6.

Fig. 1.1.7 Plots of Average Functionality and Average Dysfunctionality Points for Question J

(35)

35

From the Importance Questionnaire, the average importance is calculated for each question as in relation (1.1.7).

W

_avg

[j] =

^{∑ W}ⁱ ^ij

N (1.1.6) Represent the value near the points (Xavg [j], Yavg [j], as in figure 1.1.8. For example, draw a full circle, with the radius proportional to (Wâvg[j])^1/2, so that the radius of the circle "j" will be proportional to Wâvg [j]. As an alternative, represent the values of "j" used to identify the question in shades of gray, representing the value range for Wâvg.

Fig. 1.1.8 Plots of Average Functionality and Average Dysfunctionality Points for Question J with Average Importance Indicated for Each Point

➢ Hierarchy questionnaire

An example of a Hierarchy Questionnaire is presented in figure 1.1.9.

Fig. 1.1.9 Hierarchy questionnaire (example)

(36)

36

➢ Quality Improvement Index

It is very important for the strategy of product development, to know the quality of its product in relation to the strongest competitor. It is therefore useful not only to ask our customers to evaluate our product but also that of the competitor. The Quality Improvement Index (QI) is the ratio calculated by multiplying the relative significance of a product requirement (the Importance Questionnaire) with the difference between evaluating its product and that of the competitor through a Hierarchy Questionnaire.

QI = Relative importance x (own product evaluation - competition product evaluation).

An example of the Importance Questionnaire and Hierarchy Questionnaires (own and competitor’s) is presented in figure 1.1.10.

Fig. 1.1.10 Importance questionnaire and Hierarchy questionnaires (own and competitor’s)

1.2 HWWP, a refined IVA - Kano model

Potra, Izvercian, Pugna, and Dahlgaard (2017), proposed a refined IVA-Kano model for designing new delightful products or services. This model was first initiated in 2015.

For the last decade, companies have tried to survive in a continuous competitive global marketplace with informed and demanding customers for first-time-right delightful products and services. The present paper tries to answer the simple corporate question ‘How to design a new product for customer delight?’ by exploring

(37)

37

the relevant design requirements managers need to take into consideration for corporate strategic decision-making. After examining the ongoing debate regarding the theory of attractive quality, the Health Weapon Wealth Prospect (HWWP) model is proposed for a new product or service design, which relates Maslow’s hierarchy of needs with the Kano methodology, importance of customer wants, and the customer satisfaction coefficient. The result represents a theoretical contribution to the further development of the Kano model and a starting point for future explanatory research.

1.2.1 Introduction

Customer satisfaction is one of the key elements of a company’s financial performance and profitability (Anderson & Fornell, 1994). It is also related to the fulfillment of implicit and explicit customer needs by a product or service attributes (Tontini, Søilen, & Silveira, 2013). Transforming the voice of the customer (VOC) into design requirements for satisfying the target group may help the company capture the largest share of the market (Yang & El-Haik, 2008). In an extremely competitive global marketplace, a firm’s success resides in not only satisfying customers, but delighting them (Oliver, Rust, & Varki, 1997) for exceptional behavioral consequences such as loyalty or positive word-of-mouth, through the creation of value (Yang & Sung, 2011).

Managers cannot start the design process without having a clear image of the attributes which will differentiate and create added value for their new product or service in the marketplace. Therefore, the needs and expectations of future clients need to be listed, properly visualized, and discussed before transforming them into design requirements. This list is usually realized by applying the Kano, Seraku, Takahashi, and Tsuji (1984) model of product attributes, which influences customer satisfaction and the Kano questionnaire for classifying customer requirements.

In the last 30 years,we have witnessed an exploration and an explosion phase of research in the area of the theory of attractive quality. Witell, Löfgren, and Dahlgaard (2013) argue that further development of the Kano methodology is necessary for creating new and attractive products. In this line of reasoning, after a thorough analysis of the literature presented in the next chapter, the authors propose a theoretical contribution to the ongoing debate about the Kano methodology. The Health Weapon Wealth Prospect (HWWP) model measures customer quality attributes for a new product or service successful design in the light of customer importance of wants and

(38)

38

customer satisfaction. It visually outlines customer quality attributes situated in four different domains, which help specialists understand their position in the customer’s mind. At the same time, this proposed model represents a useful tool throughout the new product’s lifecycle. The model’s theoretical foundation is explained with the help of a case study which envisages its practical usability for business performance. The methodology, results, and discussions of the case study expose the HWWP’s implications in managerial strategic decision-making. The last section concludes the main contributions of the paper and paves the way for future research in the area of the theory of attractive quality. Theory of attractive quality debate Kano et al.’s (1984) theory of attractive quality started as an attempt to better explain the roles that different quality attributes play for customers when desiring a product or service. A methodology was constructed for theory application which classifies quality attributes in one of six quality dimensions based on customer questioning regarding the presence or absence of a product or service feature. If the feature finds itself in the A – attractive category, the customer has not thought about this characteristic but he likes the idea if it comes as a surprise. Attractive features are the ones that differentiate renowned brands from the competition. If attractive means not expected, O – one- dimensional refers to desired features the user is willing to pay for. The M – must-be category is an expected feature, the consumer assumes it as a basic requirement. The I – indifferent category represents a feature that does not influence customer satisfaction or dissatisfaction and the R – reverse category expresses a backward influence on customer satisfaction. As pointed out by Witell et al. (2013), research in this area has seen three different phases: an emergence, an exploration, and an explosion of research articles and debates regarding the Kano model. We consider ourselves remaining in the explosion phase, where several debates have taken place regarding the classification of quality attributes. Quality attributes – assignment in a category In the analysis phase of the Kano questionnaire when two categories were seen as close to one another for a single attribute, researchers developed different solutions. Matzler (1996) used a special evaluation rule M > O > A > I for a clear quality attribute category assignment. Newcomb and Lee (1997) classified the attributes in this situation as a combination, and Kano (2001) proposed a new approach regarding an attribute with two strong categories, talking for the first time about quality attributes’ dynamics. Thus, the lifecycle of a quality attribute states that it will change in time from being indifferent, to attractive, one-dimensional and

(39)

39

ultimately must-be (Lo¨fgren & Wittel, 2005). Each new Kano refined model must take into consideration this lifecycle, even when designing a new product from scratch.

Quality attributes – classified based on importance It has been observed that quality attributes affect customer satisfaction differently. Although the Kano model has many uses, Yang (2003) considered it inefficient in identifying the rate of importance for each specific attribute. Martensen and Grönholdt (2001) classified quality attributes emphasizing the importance of customer wants. The Importance–Satisfaction model (Importance–Performance Analysis (IPA)) has also identified the importance degree of each attribute as one criterion for its matrix. By keeping the importance degree customers give to specific attributes, we can understand which attributes are minimum requirements and which can be ignored. Tontini et al. (2013) argue that it is extremely important to avoid customer dissatisfaction by achieving an adequate performance of must-be attributes before offering attractive or one-dimensional attributes. Only in this manner, the one-dimensional and attractive attributes will positively affect customer satisfaction. At the same time, Moorman (2012) suggests that smart money is invested in one-dimensional and attractive features because those are the attributes that capture the hearts and minds of customers, triggering delight. The question appears:

‘How to order and understand customer quality attributes and not dissatisfy customers but delight them?’

1.2.2 Refined attractive quality models

In the attempt to answer this question, Yang (2005) has developed a refined Kano model categorizing quality in four domains (attractive, one-dimensional, must-be, and indifferent), each with two subcategories. This approach was a step closer to a Kano methodology development because the traditional six categories have been reduced to four, discarding reverse and questionable elements with no strategic importance.

Kuo, Chen, and Deng (2012) proposed a mix between the Kano model and the IPA.

Nevertheless, the IPA–Kano model has limited practical usability since the three series of attributes (must-be, one-dimensional and attractive) are introduced in a two-variable diagram (importance and performance) as circles. The circle with the highest perimeter is represented by attractive attributes, this implying that they can have higher importance than must-be attributes, which is not the case. The HWWP model for a new product or service design The refined attractive quality models discussed,

with Applications in the Theory of Attractive Quality and Six Sigma

HABILITATION THESIS

Contributions on Development of Quality Engineering and Management