5 - A Statistical Analysis of Critical Quality Tools and Companies'performance

You might also like

Download as pdf or txt
Download as pdf or txt
You are on page 1of 8

Journal of Cleaner Production 255 (2020) 120221

Contents lists available at ScienceDirect

Journal of Cleaner Production


journal homepage: www.elsevier.com/locate/jclepro

A statistical analysis of critical quality tools and companies’


performance
A. Cohen a, *, I. Alhuraish b, C. Robledo b, A. Kobi b
a
Department of Mathematics and Statistics, University of West Florida, Pensacola, Florida, 32514, USA
b
LARIS-Systems Engineering Research Laboratory, University of Angers, 49000, Angers, France

a r t i c l e i n f o a b s t r a c t

Article history: When implementing continuous improvement methodologies such as Lean and Six Sigma, one crucial
Received 7 June 2019 aspect is to determine which quality tools to implement to achieve a performance objective. The purpose
Received in revised form of this article is to determine the critical quality tools and their effects on companies’ performance. The
20 January 2020
data was collected using a questionnaire that was conducted within French industries, in Pays de la Loire
Accepted 21 January 2020
Available online 25 January 2020
region. A multiple correspondence analysis is performed first to explore and describe the data. Second,
ordinal regression models are developed to determine what are the critical quality tools that support the
Handling editor: M.T. Moreira companies’ performance in terms of quality, cost, and productivity. This case study shows that the critical
quality tools to improve quality performance would be implementing One Piece Flow and GEMBA tools.
Keywords: Next, implementing Value Stream Mapping and Takt Time tools would significantly reduce the cost
Company performance compared to not implementing those tools. The analysis shows improvement in productivity can be
Continuous improvement achieved by implementing KANBAN tool. The findings have the potential to guide companies to deter-
Ordinal regression mine which quality tools are most likely effective in improving quality, productivity, and reducing cost.
Lean six sigma
© 2020 Elsevier Ltd. All rights reserved.
Quality improvement
Multiple correspondance analysis

1. Introduction (Delgado et al., 2010; George, 2002; Alhuraish et al., 2016; Lins
et al., 2019).
Global environmental and climate change awareness has led Many research studies have investigated the critical success
companies to comply with environmental regulations to remain factors (CSFs) for the implementation of Lean and Six Sigma
competitive in the global market. Besides, companies are managing (Bortolotti et al., 2015; Netland, 2016; Knol et al., 2018; Marzag~ ao
efficiency, innovation, and operating processes for optimal perfor- and Carvalho, 2016; Ribeiro de Jesus et al., 2016). Most of the
mance. Lean manufacturing and Six Sigma are the most popular studies have tended to emphasize the critical success factors such
methodologies for achieving not only a competitive edge but also as Top management commitment and support, Education and
environment-friendly products and services (Cherrafi et al., 2016, training, Communication, Culture change, Understanding the tools
2017; García-Alcaraz et al., 2018; Caldera et al., 2017; Zhu et al., and techniques within Six Sigma methodology, and others. Suc-
2018). In fact, Lean manufacturing is a methodology that helps cessful implementation of such methodologies implies also using
companies to continuously reduce waste (e.g. overproduction, the essential quality tools that would positively impact companies’
waiting time) while Six Sigma is used to reduce the variation in performance. It is one thing to successfully implement quality tools,
processes for the ultimate goal of improving quality production and select which quality tools that would significantly improve the
performance (George, 2002; Alhuraish et al., 2017). Research organization performance is another thing. One of the challenges
studies have shown that when companies implement Lean and Six when implementing quality tools, in the Lean and Six Sigma
Sigma their performance consistently and continuously improves methodologies, is to select those tools that are the most effective for
improving the main operational performance (cost reduction,
productivity, and quality). This is because not all quality tools are
essential to a specific project. Indeed, this study, using advanced
* Corresponding author. statistical models, does show more insight about quality tools and
E-mail addresses: acohen@uwf.edu (A. Cohen), ibrahim.alhuraish@gmail.com
their companies’ performance.
(I. Alhuraish), christian.robledo@univ-angers.fr (C. Robledo), abdessamad.kobi@
univ-angers.fr (A. Kobi). Therefore, a gap has been observed in the literature in that no

https://doi.org/10.1016/j.jclepro.2020.120221
0959-6526/© 2020 Elsevier Ltd. All rights reserved.
2 A. Cohen et al. / Journal of Cleaner Production 255 (2020) 120221

published studies have conducted an investigation to examine the Pareto Diagram, Process Flow Chart, Control Chart, and regression
critical quality tools (CQTs) such as Kaizen Team, Poka-yoke, Value analysis (George, 2002; Basu, 2004; Montgomery, 2005).
Stream Mapping (VSM), Gemba, Kanban, Control Chart (CC), Takt Clegg et al. (2010) suggested that there is a great need to select
Time (TT), Design of Experiment, Cellular Layout, and many others. appropriate quality tools for a particular project. Moreover, addi-
Thus, this study focuses on this gap by determining quality tools tional research is needed regarding critical quality tools for a
that are beneficial for a particular purpose such as reducing cost, particular performance aspect, thus there is still much to learn
improving quality, or improving productivity. about critical quality tools with regard to the integration of Lean
This paper is organized as follows: Section 2 is a literature re- and Six Sigma tools for business practices. Talib et al. (2016) focused
view about critical quality tools when implementing Lean on the state of usage of quality tools in Indian service industries.
manufacturing and Six Sigma. Section 3 presents the research data Castello et al. (2019) also presented the integration of quality tools
and the methods used in this paper. The results of the study and into the ISO 9001:2008 standard in a wind power sector supply
discussion are presented in Section 4. Section 5 presents the con- chain. Most studies mentioned before did not link the importance
clusions and future research. of quality tools to a specific organization’s performance. Those
tools, if known, can be critical to ensure performance improvement
2. Literature review in a specific aspect such as quality and cost reduction. More
recently, Yadav et al. (2020) proposed a framework for enhancing
Lean manufacturing philosophy aims at improving the produc- the adoption of lean manufacturing processes as a leading frame-
tion process by minimizing wastes stemming from overproduction, work of quality improvement systems.
wait times, transport, extra processing, inventory, motion, and The quality tools, similar to the success factors, have various
defects. Thus, waste pertains to anything that is not necessary for levels of importance. It is, therefore, important that companies are
the production of a product or service (Nave, 2002; Bortolotti et al., aware of the critical quality tools that effectively impact the per-
2015; Jasti and Kodali, 2015; Garza-Reyes, 2015). Although Lean formance. This will allow the organization to design Lean Six Sigma
tools was developed in the automotive industry the concept implementation strategies with a focus on CQTs that improve a
evolved from a shop floor improvement method to become a ho- performance aspect (e.g cost reduction, quality improvement).
listic approach that aims at creating value for stakeholders (Hines There is a clear lack of literature on the critical quality tools
et al., 2004; Mathaisel, 2005). It is also shown that a focus on compared to critical success factors. To the best of our knowledge
value creation for customers is an important element of Lean tools there is no published studies that investigated the critical quality
because the same outcomes are achieved by reducing costs of tools such as Kaizen Team, Poka-yoke, Value Stream Mapping,
suppliers and avoiding waste, hence ensuring the effective use of Gemba, Kanban, Control Chart, Takt Time, Design of Experiment,
resources (Hines et al., 2004). Shah and Ward (2007) explained that Cellular Layout, and others, and their impact on organizations’
Lean tools production is a socio-technical system that avoids waste performance. Thus, this study focuses on this gap by determining
by simultaneously reducing supplier, customer, and internal quality tools that are beneficial for a particular performance pur-
variability. pose such as reducing cost, improving quality, and improving
Six Sigma approach deals with improving efficiency by identi- productivity.
fying deficiencies in existing processes. It deals also with imple-
menting and controlling a modified or new process to improve 3. Material and methods
customer value and process efficiency. Researchers describe Six
Sigma tools as a data-driven approach to problem-solving that in- A survey was designed to gather data on the implementation of
corporates statistical methods into management (Blakeslee Jr, Lean and Six Sigma tools within companies in Pays de La Loire,
1999; Braunscheidel et al., 2011). Pyzdek and Keller (2014) France. This may be the first study of its kind in France. The survey
considered it to be a rigorous approach to improve quality in or- participants were also required to respond to questions by indi-
ganizations because of the use of standardized tools and methods. cating whether or not they agreed with a statement with a certain
Furthermore, studies have documented various benefits of imple- level of agreement or disagreement as follows, Strongly Disagree
menting Six Sigma tools such as an increase in savings, increase in (1), Disagree (2), Neutral (3), Agree (4), Strongly Agree (5). A score
perceived innovation (Braunscheidel et al., 2011), improvements in of (5) represents the highest value of agreement and (1) represents
knowledge transfer and organizational learning (Choo et al., 2007), the maximum value of disagreement. The response variables have a
and improvements in overall quality of products or services (Levine Likert scale then.
and Toffel, 2010). Recently, Lean and Six Sigma methodologies have The goal is to investigate the impact of the implementation of
been associated also with concepts such as eco-efficiency, sus- Lean and Six Sigma tools on the companies’ performance in terms
tainability, and Green (Moreira et al., 2010; Maia et al., 2013; Abreu of three criteria, quality improvement (Q), productivity (P), and cost
et al., 2017; Garcia-Alcaraz et al., 2018; Amaro et al., 2019). Lean reduction (C). The data collected are used to develop regression
tools have improved financial and environmental performance of models where each criterion is the response variable of a model,
the organization, which also helped to achieve sustainability thus three models are developed. Thirty-three companies
(Henao et al., 2019; Lermen et al., 2018; Abu et al., 2019). completed the survey. Various sectors of industries and businesses
Successful implementation of Lean and Six Sigma approaches is completed the questionnaire, including 10 companies from the
driven by successful application of quality tools. Those widely used electronics industry, 7 companies from automotive industry, 4
in Lean manufacturing include Kaizen, Kanban, Brainstorming, companies from the health industry, 2 companies from the trans-
Value stream Map (VSM), 5S, Total productivity Maintenance portation industry, 6 companies from service, and 4 companies are
(TPM), Kaizen team, Work cell optimization, Mistake proofing & from other industry such as energy, oil and gas, food and aerospace
prevention (Poka-yoke), Gemba, PDCA, Standardized Work, Takt industries. Fourteen quality tools (independent variables) are
time, Visual Control, Once Piece Flow, and Single Minute Exchange included in this study. Table 1 shows the percent of the imple-
of Die (SMED). On the other side, Six Sigma often uses statistical mentation of each quality tool within our sample of 33 companies.
tools such as Voice of customer (VOC), Cause and Effect Diagram, For example, Regression analysis as a quality tool is only imple-
Check Sheet, Design of Experiment (DOE), Defect Per Million Op- mented within 13 companies. Three criteria are given to assess the
portunities (DPMO), Failure Mode and Effects Analysis (FMEA), performance of the companies (dependent variables).
A. Cohen et al. / Journal of Cleaner Production 255 (2020) 120221 3

Table 1
Independent variables (quality tools) considered in the analysis.

Variables Type Percent of implementation in the sample

- Kaizen Team Categorical; 1 ¼ not implemented, 2 ¼ implemented 81.8%


- Poka-yoke Categorical; 1 ¼ not implemented, 2 ¼ implemented 81.8%
- Value Stream Mapping (VSM) Categorical; 1 ¼ not implemented, 2 ¼ implemented 81.8%
- Gemba Categorical; 1 ¼ not implemented, 2 ¼ implemented 78.8%
- Kanban Categorical; 1 ¼ not implemented, 2 ¼ implemented 78.8%
- Single-Minute Exchange of Dies (SMED) Categorical; 1 ¼ not implemented, 2 ¼ implemented 72.7%
- Control Chart (CC) Categorical; 1 ¼ not implemented, 2 ¼ implemented 72.7%
- Takt Time (TT) Categorical; 1 ¼ not implemented, 2 ¼ implemented 69.7%
- Design of Experiment Categorical; 1 ¼ not implemented, 2 ¼ implemented 63.6%
- Defects Per Million Categorical; 1 ¼ not implemented, 2 ¼ implemented 60.6%
Opportunities (DPMO)
- One Piece Flow (OPF) Categorical; 1 ¼ not implemented, 2 ¼ implemented 57.6%
- Define, Measure, Analyze, Categorical; 1 ¼ not implemented, 2 ¼ implemented 57.6%
Improve and Control (DMAIC)
- Cellular Layout (CellLay) Categorical; 1 ¼ not implemented, 2 ¼ implemented 48.5%
- Regression Analysis (Reg) Categorical; 1 ¼ not implemented, 2 ¼ implemented 39.4%

Fig. 1 presents the performance perceived by the companies 3.2. Ordinal logistic regression
through the Likert scale. For instance, most of the companies has
received a satisfactory quality improvement (Q) since 19 companies Many different regression models for analyzing ordinal
responded Agree (4) and 12 answered Strongly Agree (5), a total of response variables were being developed during the past decades
31 out of 33. (Harrell Jr, 2015; McCullagh, 1980; McKelvey and Zavoina, 1975).
Ordinal regression model is included in the general framework of
generalized linear models (GLM), where various models may result
from the use of different link functions (logistic, loglog, probit). The
most commonly used ordinal regression model is the proportional
3.1. Multiple correspondence analysis
odds (PO) model (McCullagh, 1980). Let’s consider a response var-
iable Y having levels 1; 2; …; k, the logit ordinal regression equation
Multiple correspondence analysis (MCA) allows us to study the
is given as follows:
relationships of several categorical variables. It can be seen as a
generalization of principal component analysis (PCA) when the  
PðY  jjXÞ
variables are categorical instead of continuous (Benze cri, 1977; logit½PðY  jjXÞ  ¼ log ¼ aj þ bX; (4)
1  PðY  jjXÞ
Abdi and Valentin, 2007). It is useful to describe, explore, sum-
marize, and visualize information contained within a data table of I where j2f1; 2; …; k  1g. For a given level j (say j ¼ 2 corresponds
individuals described by M categorical variables. It is an extension to “Disagree” in our survey), the model is a logistic regression for
of correspondence analysis (CA) to the case of more than two cat- the event Y  j and has its own intercept. A crucial assumption for
egorical variables (Husson and Josse, 2014). Consider each cate-
P the proportional odds model is the parallel regression assumption,
gorical variable has Jq levels and Mq¼1 Jq ¼ J. Given I individuals, the where it is assumed to have the binary logits with identical slopes
I  J indicator matrix is denoted X. Computing a correspondence and different intercepts. To check whether or not the parallel
analysis on the indicator matrix provides two sets of factors scores regression assumption holds, one can use the Likelihood ratio test,
(rows and columns). The proportion matrix P is then calculated, by the graphical diagnostics, or the Brant test (Brant, 1990).
dividing the element of X by the total of all number in X, and
denoted P ¼ Xij =nX . The total rows and the columns of P are equal
to r ¼ P1 and c ¼ P’1, respectively; where 1 is an appropriately 4. Results and discussion
dimensioned vector of ones. Let’s define the two matrices Dr ¼
diagðrÞ1=2 and Dc ¼ diagðcÞ1=2 . The factor scores are obtained This investigation starts by applying the multiple correspon-
from the singular value decomposition, as follows: dence analysis to the quality tools. “The Design of Experiment” tool
was excluded because of its high association with the “Control
  Chart” tool (Spearman’s rank correlation ¼ 0.81, p-value
Dr P  rcT Dc ¼ Z DQ T (1) < 0:00001). The data matrix consists of 33 rows (companies) and 13
columns (quality tools).
where D is the diagonal matrix of the singular values. Z and Q are Fig. 2 (left) shows the biplot of individuals (companies pre-
matrices from the singular value decomposition. The row and col- sented by numbers) and variables categories into the first two di-
umns factor scores are given as follows, respectively: mensions of the MCA. The percentage of variability explained into
the two dimensions is equal to 55.9% ¼ 39.9% (Dim1) þ 16% (Dim2).
F ¼ Dr Z D (2) In the biplot, the distance between any points (companies or var-
iable categories) gives a measure of similarity that shows a general
pattern in the data. For example, companies 13, 8, 5, and 3 form an
G ¼ Dc Q D (3)
isolated cluster from the other companies, showing that they have
The MCA helps in exploring the data by providing a typology of overall similar answers to the questions, which means these com-
the companies and the similarities between the companies from a panies are using similar quality tools, such as Control Chart (CC) and
multidimensional perspective. In addition to assessing the re- SMED, but not all of them. In addition, this cluster is far from the
lationships within the quality tools being implemented in regard to center of the biplot, which means that these companies forming the
the companies. cluster are not implementing most of the quality tools. The closer
4 A. Cohen et al. / Journal of Cleaner Production 255 (2020) 120221

Fig. 1. Companies’ performance reported in the survey, based on the implementation of the quality tools, in terms of Quality (Q), Cost (C), and Production (P).

the companies are to the center the more quality tools they companies being questioned, as well as Reg (Regression analysis),
implement. A company that exactly falls in the center would DPMO, and CellLay (Cellular Layout). On the other side, there is a
implement all quality tools being investigated. Note that most clear discriminative pattern regarding companies that whether or
variables are well presented by the two dimensions as shown in not use GEMBA. An identical pattern can be recognized for VSM.
Fig. 2 (right), that is the squared correlations between the variables Furthermore, looking at the ellipses in Fig. 3, one can observe
and each dimension are used as coordinates. The variables Poka- similar patterns such as for GEMBA-VSM, Kaisen-KANBAN, and
yoke (Poka) and Takt Time (TT) are the most correlated variables OPF-TT-SMED. These similarities indicate that those variables
with dimension 1. Similarly, DMAIC and SMED are the most (quality tools) are whether or not implemented within the same
correlated with dimension 2. It can be seen that the variables companies. For instance, the companies that implement GEMBA
GEMBA and VSM are well presented in both dimensions (1e2). will most likely implement VSM and vice-versa. From this explor-
Fig. 3 presents the individuals (companies) by groups using the atory statistical analysis, It is expected that tools such as GEMBA,
levels of the variables (1 ¼ not implemented, 2 ¼ implemented). A KANBAN, and VSM would have a significant impact on the com-
95% confidence ellipse is used around the mean point of each group panies’ performance.
assuming a multivariate t-distribution. This result helps to indicate Next, thirteen tools are used as factors in the ordinal regression
which variables clearly separate the two groups of companies models where quality (Q), cost reduction (C), and productivity (P)
(1 ¼ not implemented, 2 ¼ implemented). For example, the control are the response variables. The ordinal logistic regression is used to
chart (CC) tool does not provide a discriminative feature for the study the relationship between the tools and the performance. A

Fig. 2. The multiple correspondence analysis.


A. Cohen et al. / Journal of Cleaner Production 255 (2020) 120221 5

likelihood ratio test was performed to make sure that the parallel KANBAN and SMED, in a positive direction.
regression assumption is met. Further, Table 3 provides the reduced models where all the
The estimated coefficients of the predictors in the ordinal lo- predictors are statistically significant. The Likelihood Ratio test
gistic regression models are given in Table 2. There are three compares the log likelihoods of the two models (full vs. reduced), if
models one for each response variable. Bold typeface values indi- this difference is statistically significant, then the model with more
cate statistically significant predictors (p-value < 0:1Þ. For the variables is said to fit the data significantly better than the more
Quality (Q) full model, two tools are statistically significant, that are restrictive model. The resulting test statistic is distributed chi-
OPF and DPMO. In terms of the sign/direction of the coefficients. squared, with degrees of freedom equal to the number of param-
The OPF estimate is positive, which means when OPF is “imple- eters that are constrained.
mented” the company’s statement about quality improvement is The reduced model of Quality (Q) shows a significant association
more likely to be “Strongly Agree” than companies that do not between GEMBA, OPF, and DPMO implementation and how likely
implement OPF. On the other side, the DPMO estimate is negative, the company will perceive a better quality performance. The odds
which tells us that when the DPMO tool is not implemented the ratios can help interpret the result. For instance, a company that
company response about the quality improvement is most likely to goes from not implement (¼1) to implement (¼2) GEMBA tool has
be “Disagree”. Regarding the Cost model (C), the only significant odds of answering “Strongly Agree”, about the quality improve-
predictor is Cellular Layout tool and has a negative estimate. The ment versus answering “Neutral” and “Agree”, 9.216 greater, given
Productivity model (P) has two significant predictors, that are that all other variables in the model are held constant. The same

Fig. 3. 95% Confidence ellipses grouping companies based on the variables categories (1 ¼ not implemented, 2 ¼ implemented) using a multivariate t-distribution.
6 A. Cohen et al. / Journal of Cleaner Production 255 (2020) 120221

Table 2
Full ordinal logistic regression models (10% is level of significance).

Predictors Ordinal Logistic Regression Model with Response Variable

Quality Cost Productivity

Estimate SE P-value Estimate SE P-value Estimate SE P-value

VSM 0.1391 3.3410 0.9668 2.9128 1.9626 0.13777 0.99221 2.13659 0.6424
GEMBA 4.3119 2.9532 0.1443 1.9241 1.5879 0.22560 1.29724 1.62342 0.4242
Poka 0.7509 1.7763 0.6725 0.5739 2.0955 0.78419 1.45231 1.89129 0.4426
KANBAN 2.2269 2.2476 0.3218 2.5199 1.9069 0.18635 4.32320 2.00322 0.0309
TT 0.5339 1.5553 0.7314 1.4046 1.4864 0.34468 1.32455 1.35590 0.3286
Kaizen 0.5907 1.8844 0.7539 2.2181 1.8944 0.24164 1.94919 1.65009 0.2375
SMED 0.8911 1.5016 0.5529 2.2127 1.5934 0.16493 2.33915 1.41836 0.0991
DMAIC 1.6513 1.7101 0.3342 0.1337 1.3508 0.92114 1.46815 1.48276 0.3221
OPF 3.8409 2.2000 0.0808 0.1352 1.7253 0.93752 0.38800 1.53841 0.8009
Reg 1.9378 1.5975 0.2251 0.3629 1.3100 0.78175 1.60543 1.37059 0.2415
CellLay 1.3559 1.3021 0.2977 3.7996 1.6240 0.01930 1.58243 1.29882 0.2231
DPMO 4.0665 1.9158 0.0338 0.2186 1.3847 0.87457 0.32822 1.22352 0.7885
CC 1.6058 1.3029 0.2178 1.6291 1.2917 0.20726 1.15690 1.21825 0.3423

Log-likelihood 20.5025 25.3933 28.5941


AIC 73.0051 82.7867 89.1881

thing for OPF with an odds ratio of 13.22. On the other side, DPMO likely will increase the odds of improving their performance met-
shows an odds ratio less than one, which means the odds of rics. Consequently, more research should be conducted to investi-
answering “Strongly Agree” about the quality improvement versus gate the critical quality tools that organizations should implement
answering “Neutral” and “Agree” decreases with a multiple of ð1  effectively and efficiently to achieve a performance objective.
0:055Þ ¼ 0:945. This study explored 13 quality tools or techniques related to
The reduced model Cost (cost reduction) has three predictors, Lean manufacturing and Six Sigma. An exploratory analysis was
that are CellLay, VSM, and TT. The odds ratios are given in Table 3. conducted using multiple corresponding analysis to discover the
For VSM (Value Stream Mapping) tool, the odds of a company similarities among companies, and also among quality tools. The
answering “Strongly Agree” about reducing the cost versus data used in this study was ordinal. Therefore, an ordinal regression
answering “Neutral” and “Agree” combined are 13.503 greater, model was developed to investigate the impact of every single
when all other variables in the model are held constant and the quality tool on a particular performance goal. The findings show
VSM goes from not implemented to implemented. A similar that 1) the critical quality tool to improve quality performance
conclusion is given for TT (Takt Time) tool with an odds ratio of would be implementing One Piece Flow and GEMBA tools, 2)
10.697. Regarding CellLay tool the odds ratio is less than one. implementing Value Stream Mapping and Takt Time tools would
The Productivity model shows that KANBAN tool has a big role significantly reduce the cost within a company compared to not
to improving productivity. The odds ratio is 10.777, which means a implementing these tools, and 3) An improvement in productivity
company that implements KANBAN would show “Strongly Agree” can be achieved by implementing the KANBAN tool. The findings
answer about productivity improvement versus “Neutral” and can be useful to guide companies to direct their resources or efforts
“Agree” combined. The increase is a multiple of 10.777. to the critical quality tools instead of implementing all the tools
It is crucial to be aware that not all organizations, and in equally.
particular small and medium enterprises (SMEs), have the capacity It is worth noting that generalization would be difficult to
to implement quality tools, simultaneously or as needed, because of consider outside France because of the nature of the sample. Next,
shortage of funds, expertise, or time. This is important because this future work should definitely focus on collecting a larger sample or
goes along with which critical quality tools that should be imple- building a new case study to support the statistical analysis and
mented to achieve a specific performance goal. Indeed, organiza- findings of this study. There is a huge lack in the literature about
tions will focus on implementing those CQTs, if known, that most critical quality tools and their impact on performance compared to

Table 3
Reduced ordinal logistic regression models (10% is level of significance).

Predictors Reduced Ordinal Logistic Regression Models with Response Variable

Estimate SE P-value Odds Ratio P-value Likelihood Ratio Test

Quality (Full vs. Reduced)

GEMBA 2.221 1.283 0.0835 9.216 0.891


OPF 2.582 1.321 0.0501 13.22
DPMO 2.895 1.314 0.0276 0.055

Cost

CellLay 2.08366 0.9046 0.0213 0.1244 0.702


VSM 2.60298 1.1762 0.0269 13.503
TT 2.36993 1.0261 0.0209 10.697

Productivity

KANBAN 2.37742 0.89971 0.0082 10.777 0.5695


A. Cohen et al. / Journal of Cleaner Production 255 (2020) 120221 7

critical success factors. We believe future studies should include sustainable business practice: a systematic literature review. J. Clean. Prod. 167,
1546e1565.
more factors such as the nature of the project, timeline for the
Castello, J., De Castro, R., Marimon, F., 2019. Use of quality tools and techniques and
implementation, and any other factors that may contribute to their integration into iso 9001. Int. J. Qual. Reliab. Manag. 37 (1), 68e89. https://
improve the performance. It is important to account for most fac- doi.org/10.1108/IJQRM-07-2018-0171.
tors that may be confounded with the effects of the critical quality Cherrafi, A., Elfezazi, S., Chiarini, A., Mokhlis, A., Benhida, K., 2016. The integration of
lean manufacturing, six sigma and sustainability: a literature review and future
tools such as employees incentives initiatives. This work could be research directions for developing a specific model. J. Clean. Prod. 139,
extended to include the cost of implementing the quality tools as 828e846.
well their environmental impacts. Cherrafi, A., Elfezazi, S., Govindan, K., Garza-Reyes, J.A., Benhida, K., Mokhlis, A.,
2017. A framework for the integration of green and lean six sigma for superior
sustainability performance. Int. J. Prod. Res. 55 (15), 4481e4515.
5. Conclusions Choo, A.S., Linderman, K.W., Schroeder, R.G., 2007. Method and psychological effects
on learning behaviors and knowledge creation in quality improvement projects.
Manag. Sci. 53 (3), 437e450.
This paper investigated the critical quality tools using a survey Clegg, B., Rees, C., Titchen, M., 2010. A study into the effectiveness of quality
to question french companies about their performance and the management training: a focus on tools and critical success factors. The TQM
status of implementation of quality tools. The ordinal regression Journal 22 (2), 188e208.
Delgado, C., Ferreira, M., Castelo Branco, M., 2010. The implementation of lean six
models being developed helped understand the importance of sigma in financial services organizations. J. Manuf. Technol. Manag. 21 (4),
some of the quality tools, which we called critical quality tools. A 512e523.
sample of 33 companies was obtained. The findings showed that García-Alcaraz, J., Alor-Hern andez, G., Sa nchez-Ramírez, C., Jime nez-Macías, E.,
Blanco-Fern andez, J., Latorre-Biel, J., 2018. Mediating role of the six sigma
critical quality tools to improve quality performance would be implementation strategy and investment in human resources in economic
implementing One Piece Flow and GEMBA tools. Furthermore, success and sustainability. Sustainability 10 (6), 1828.
implementing Value Stream Mapping and Takt Time tools would Garcia-Alcaraz, J.L., Alor-Hern andez, G., Sa nchez-Ramírez, C., Jimenez-Macías, E.,
Blanco-Fern andez, J., Latorre-Biel, J.I., 2018. Mediating role of the six sigma
significantly reduce the cost within a company compared to not
implementation strategy and investment in human resources in economic
implementing these tools. The analysis suggested also that an success and sustainability. Sustainability 10 (6), 1828.
improvement in productivity can be achieved by implementing the Garza-Reyes, J.A., 2015. Lean and greenea systematic review of the state of the art
KANBAN tool. Therefore, a recommendation to the companies literature. J. Clean. Prod. 102, 18e29.
George, M., 2002. Lean Six Sigma: Combining Six Sigma Quality with Lean Pro-
would be to making sure the effective implementation of these duction Speed. McGraw Hill Professional.
critical quality tools and all resources should be made available to Harrell Jr., F.E., 2015. Regression Modeling Strategies: with Applications to Linear
achieve a successful implementation. Models, Logistic and Ordinal Regression, and Survival Analysis. Springer.
Henao, R., Sarache, W., Go  mez, I., 2019. Lean manufacturing and sustainable per-
formance: trends and future challenges. J. Clean. Prod. 208, 99e116.
Contributions of authors Hines, P., Holweg, M., Rich, N., 2004. Learning to evolve: a review of contemporary
lean thinking. Int. J. Oper. Prod. Manag. 24 (10), 994e1011.
Husson, F., Josse, J., 2014. Multiple Correspondence Analysis. Visualization and
Achraf Cohen: Conceptualization, Methodology, Software, Data Verbalization of Data, pp. 165e184.
curation, Writing e Original Draft and Reviewing and Editing, Jasti, N.V.K., Kodali, R., 2015. Lean production: literature review and trends. Int. J.
Ibrahim Alhuraish: Conceptualization, Investigation, Data curation, Prod. Res. 53 (3), 867e885.
Knol, W.H., Slomp, J., Schouteten, R.L., Lauche, K., 2018. Implementing lean practices
Writing - Reviewing and Editing, Abdessamad Kobi: Supervision. in manufacturing smes: testing ‘critical success factors’ using necessary con-
Christian Robeldo: Supervision. dition analysis. Int. J. Prod. Res. 56 (11), 3955e3973.
Lermen, F.H., Echeveste, M.E., Peralta, C.B., Sonego, M., Marcon, A., 2018.
A framework for selecting lean practices in sustainable product development:
Declaration of competing interest the case study of a brazilian agroindustry. J. Clean. Prod. 191, 261e272.
Levine, D.I., Toffel, M.W., 2010. Quality management and job quality: how the iso
None. 9001 standard for quality management systems affects employees and em-
ployers. Manag. Sci. 56 (6), 978e996.
Lins, M.G., Zotes, L.P., Caiado, R., 2019. Critical factors for lean and innovation in
References services: from a systematic review to an empirical investigation. Total Qual.
Manag. Bus. Excel. 1e26.
Abdi, H., Valentin, D., 2007. Multiple Correspondence Analysis. Encyclopedia of Maia, L.C., Alves, A.C., Lea ~o, C.P., 2013. Sustainable work environment with lean
measurement and statistics, pp. 651e657. production in textile and clothing industry. International Journal of Industrial
Abreu, M.F., Alves, A.C., Moreira, F., 2017. Lean-green models for eco-efficient and Engineering and Management 4 (3), 183e190.
sustainable production. Energy 137, 846e853. Marzaga ~o, D.S.L., Carvalho, M.M., 2016. Critical success factors for six sigma projects.
Abu, F., Gholami, H., Saman, M.Z.M., Zakuan, N., Streimikiene, D., 2019. The Int. J. Proj. Manag. 34 (8), 1505e1518.
implementation of lean manufacturing in the furniture industry: a review and Mathaisel, D.F., 2005. A lean architecture for transforming the aerospace mainte-
analysis on the motives, barriers, challenges, and the applications. J. Clean. Prod. nance, repair and overhaul (mro) enterprise. Int. J. Prod. Perform. Manag. 54 (8),
234, 660e680. https://doi.org/10.1016/j.jclepro.2019.06.279. 623e644.
Alhuraish, I., Robledo, C., Kobi, A., 2016. Impacts of lean manufacturing and six McCullagh, P., 1980. Regression models for ordinal data. J. Roy. Stat. Soc. B 109e142.
sigma. In: 22nd ISSAT International Conference on Reliability and Quality in McKelvey, R.D., Zavoina, W., 1975. A statistical model for the analysis of ordinal level
Design-RQD. dependent variables. J. Math. Sociol. 4 (1), 103e120.
Alhuraish, I., Robledo, C., Kobi, A., 2017. A comparative exploration of lean Montgomery, D., 2005. Introduction to Statistical Quality Control. John Wiley,
manufacturing and six sigma in terms of their critical success factors. J. Clean. Hoboken, N.J.
Prod. 164, 325e337. Moreira, F., Alves, A.C., Sousa, R.M., 2010. Towards eco-efficient lean production
Amaro, P., Alves, A.C., Sousa, R.M., 2019. Lean Thinking: A Transversal and Global systems. In: International Conference on Information Technology for Balanced
Management Philosophy to Achieve Sustainability Benefits. Springer Interna- Automation Systems. Springer, pp. 100e108.
tional Publishing, Cham, pp. 1e31. Nave, D., 2002. How to compare six sigma, lean and the theory of constraints. Qual.
Basu, R., 2004. Implementing Quality: a Practical Guide to Tools and Techniques: Prog. 35 (3), 73.
Enabling the Power of Operational Excellence. Cengage Learning EMEA. Netland, T.H., 2016. Critical success factors for implementing lean production: the
Benzecri, J.-P., 1977. Sur l’analyse des tableaux binaires associe s a
 une corre- effect of contingencies. Int. J. Prod. Res. 54 (8), 2433e2448.
spondance multiple. CAD 2, 55e71. Pyzdek, T., Keller, P.A., 2014. The Six Sigma Handbook. McGraw-Hill Education, New
Blakeslee Jr., J.A., 1999. Implementing the six sigma solution. Qual. Prog. 32 (7), 77. York.
Bortolotti, T., Boscari, S., Danese, P., 2015. Successful lean implementation: organi- Ribeiro de Jesus, A., Antony, J., Lepikson, H.A., Peixoto, A.L., 2016. Six sigma critical
zational culture and soft lean practices. Int. J. Prod. Econ. 160, 182e201. success factors in brazilian industry. Int. J. Qual. Reliab. Manag. 33 (6), 702e723.
Brant, R., 1990. Assessing proportionality in the proportional odds model for ordinal Shah, R., Ward, P.T., 2007. Defining and developing measures of lean production.
logistic regression. Biometrics 1171e1178. J. Oper. Manag. 25 (4), 785e805.
Braunscheidel, M.J., Hamister, J.W., Suresh, N.C., Star, H., 2011. An institutional Talib, F., Rahman, Z., Qureshi, M., 2016. Survey on the usage of total quality man-
theory perspective on six sigma adoption. Int. J. Oper. Prod. Manag. 31 (4), agement tools and techniques in indian service industries: an empirical anal-
423e451. ysis. Int. J. Qual. Innovat. 2 (2), 105e119. Talib, F., Rahman, Z. and Qureshi, MN
Caldera, H., Desha, C., Dawes, L., 2017. Exploring the role of lean thinking in (2013)’Survey on the usage of total quality management tools and techniques in
8 A. Cohen et al. / Journal of Cleaner Production 255 (2020) 120221

Indian service industries: an empirical analysis’. 118726.


Yadav, G., Luthra, S., Huisingh, D., Mangla, S.K., Narkhede, B.E., Liu, Y., 2020. Zhu, Q., Johnson, S., Sarkis, J., 2018. Lean six sigma and environmental sustain-
Development of a lean manufacturing framework to enhance its adoption ability: a hospital perspective. In: Supply Chain Forum: an International Journal,
within manufacturing companies in developing economies. J. Clean. Prod. 245, vol. 19. Taylor & Francis, pp. 25e41.

You might also like