Statistical Analyses and Modeling of the Implementation of Agile Manufacturing Tactics in Industrial Firms

This paper provides a review and introduction on agile manufacturing. Tactics of agile manufacturing are mapped into different production areas eight-construct latent : manufacturing equipment and technology, processes technology and know-how, quality and productivity improvement, production planning and control, shop floor management, product design and development, supplier relationship management, and customer relationship management. The implementation level of agile manufacturing tactics is investigated in each area. A structural equation model is proposed. Hypotheses are formulated. Feedback from 456 firms is collected using five-point-Likert-scale questionnaire. Statistical analysis is carried out using IBM SPSS and AMOS. Multicollinearity, content validity, consistency, construct validity, ANOVA analysis, and relationships between agile components are tested. The results of this study prove that the agile manufacturing tactics have positive effect on the overall agility level. This conclusion can be used by manufacturing firms to manage challenges when trying to be agile.


Introduction
Agile manufacturing AM is described as new tactics of manufacturing.It emerged after lean production LP .It represents pattern shifts from mass production MP .It originated from the 21st century manufacturing enterprise study that was conducted at Lehigh University in the early 1990s 1 .Following that, a book entitled "Agile Companies and Virtual Enterprise" recognized as the state-of-the-art work on AM was published in 1995.
According to Groover 1 "agile manufacturing can be defined as: 1 an enterprise level manufacturing strategy of introducing new products into rapidly changing markets, 2 an organizational ability to thrive in a competitive environment characterized by continuous and sometimes unforeseen change".Pham et al. 2 defined agile manufacturing as the ability

Agile Selection of Manufacturing Equipment and Technology (MET)
The relations between the specific equipment configurations with visual control and group technology are developed by 5, 6 .The explanations of how to design cellular layouts are given by 7, 8 .Li-Hua and Khalil 9 investigated the rapid changes in the business environment.They showed how companies can maximize business opportunities when the risks are considered.The relation between new equipment/technologies and production process reengineering is developed by 10, 11 .A discussion about how the production process reengineering increases productivity and efficiency is presented in 12, 13 .In order to have a formal investigation about the effect of MET, the following hypotheses are proposed: H1 0 : MET implementation has a significant, positive effect on the development of AMS.
H1 1 : MET implementation has no effect on the development of AMS.
H2 0 : MET implementation has a significant, positive effect on the development of PTK.
H2 1 : MET implementation has no effect on the development of PTK.
H3 0 : MET implementation has a significant, positive effect on the development of QPIM.
H3 1 : MET implementation has no effect on the development of QPIM.
H4 0 : MET implementation has a significant, positive effect on the development of SFM.
H4 1 : MET implementation has no effect on the development of SFM.
H5 0 : MET implementation has a significant, positive effect on the development of PDD.
H5 1 : MET implementation has no effect on the development of PDD.

Processes Technology and Know-How (PTK)
The elimination of waste can 1 simplify organizations processes 14 , 2 allow business to be more agile and dynamic which offers the opportunity to meet customer demands in new products and services, and 3 allow business to be more responsive to customers' concerns 15 .
Researchers on AM have established that flexibility is the foundation of AM.Flexibility is classified into machine flexibility, routing flexibility, product flexibility, manufacturing system flexibility, strategic flexibility, volume flexibility, and so forth, 16 .Yusuf et al. 14 stated that "agility is reflected in: the successful exploration of competitive bases through the integration of reconfigurable resources and best practices in a knowledge-rich environment to provide customer-driven products and services in a fast-changing market environment".
Quinn et al. 17 defined agility as the ability to accomplish rapid changeover from the assembly of one product to the assembly of different product 18 .Gunasekaran 19 found that the rapidprototyping is one of the major enablers of agility.Prototyping describes the design and generation of an early version of product.Many strategies/techniques such as rapid-partnership formation, e-manufacturing, and rapid prototyping can be employed to improve the responsiveness of the overall system for customer requirements 19 .This eventually leads to an increase in the customers' investments.Accordingly, the following hypotheses are formulated: H6 0 : PTK implementation has a significant, positive effect on the development of AMS.

Quality and Productivity Improvement and Measures (QPIM)
Hormozi 20 stated that agile manufacturing produces defect free product.Misra et al.
of in-process defects.Agility achieves improvements in productivity and quality through flexibility of access and utilization of resources 22 .Gunasekaran 19 showed that manufacturing performance measures such as productivity would help to design the most effective agile manufacturing system.
Agile-based manufacturing organizations have higher productivity market shares 23 .Several researchers use productivity and quality as measures for process performance 24 .Others use different measures such as the multidimensional index created by Schroeder et al. 25 .AM requires modular production facilities.Gunasekaran 26 found out that AM characterized the needs for modular production facilities in decision making.
AM involves fundamental change in an organization's approach to cycle-time reduction 27 .Naylor et al. 28 showed the necessity for production lead time reduction as a prerequisite to agility.Short production lead times were addressed in 23 .Sieger et al. 29 measured responsiveness of companies relative to the product development cycle time.Vinodh and Kuttalingam 30 suggested that the one major constituent of AM is the minimization of manufacturing lead times.Accordingly, the following hypotheses are proposed: H8 0 : QPIM implementation has a significant, positive effect on the development of AMS.

Production Planning and Control (PPC)
Production planning and control PPC plays an important role in the competitive environments.PPC responds immediately to achieve higher service level of performance, better resource utilization, and less material loss.Yan 31 established an approach to stochastic production planning SPP for flexible automation in agile manufacturing environment.Li et al. 32 concluded that the performance of customer service level in enterprises is highly dependent on the effectiveness of its manufacturing planning and control system.Chen 33 discussed four problems of production management in the environment of agile manufacturing.These problems are 1 organization of production, 2 production planning, 3 production control, and 4 quality control.Le et al. 34 described the production planning methodology that can be implemented in agile manufacturing.They studied two multiitem lot-sizing problems.They detailed the development of the planning problem mathematically and highlighted solutions to some of their initial problems.Tunglun and sato 35 provided a model of PPC that concretely defines the PPC and allows the possibility for immediate planning and scheduling.
Gold and Thomas 36 discussed and simulated lean, agile, and hybrid supply chain strategies.Their study demonstrated that while lean management typically calls for maketo-stock replenishment driven by short-term forecasts, agile supply chains employ make-toorder provisions.Ching et al. 37 provided a structured procedure for identifying the agile drivers in the business environment.They determined the management information system requirements that enhance manufacturing agility.Adrian et al. 38 studied the evolution of information systems in manufacturing and its importance in supporting agile manufacturing.Lenny and Mike 39 concluded that the application of enterprise resource planning ERP has improved agility and responsiveness.Petri 40 showed that resource management is an important part of any production system, especially when building agility in the manufacturing of the company.David and Chong 41 presented a review of agile supply partner decision making published between 2001 and 2011.The progress made in developing new models and methods applicable to this task is assessed in the context of the previous literature.Particular attention is given to those methods that are especially relevant for the use of agile in supply chains.The review highlighted the ongoing need for developing methods that are able to meet the combination of qualitative and quantitative objectives.These objectives are typically found in partner selection.Based on previous discussion, we theorized the following hypotheses: H11 0 : PPC implementation has a significant, positive effect on the development of AMS.

Shop Floor Management (SFM)
In 1995, shop floor control functional diagram was developed by Technologies Enabling Agile Manufacturing "TEAM" 42 .A hybrid integration approach was developed to solve the problem of shop floor scheduling 43 .Ribeiro et al. demonstrated how the seamless integration of the shop floor with external tools is achieved 44 .A multiagent architecture of agile manufacturing system and a hybrid strategy for shop floor scheduling were adopted by Li et  H13 0 : SFM implementation has a significant, positive effect on the development of AMS.
H13 1 : SFM implementation has no effect on the development of AMS.

Product Design and Development (PDD)
Andrew 53 considered the outsourcing strategy and how it affects product design.He explained how outsourcing permits manufacturers to remain more agile and competitive by retaining local manufacture.Computer-aided design CAD is used to bring out new models for achieving design agility 54 .Agility is greatly influenced by the emergence and growth of new technologies such as CAD, CAM, CNC, RP, and so forth, 55 .Modular architecture for developing product platform is crucial to agile manufacturing and product variety that satisfies various customers' needs and high agility 56 .The relations of CAPP/CAM packages, simulators, design analysis and synthesis tools, and decision support systems with agility are discussed in 57 .Based on this analysis, we postulate the following hypothesis: H14 0 : PPD implementation has a significant, positive effect on the development of AMS.
H14 1 : PPD implementation has no effect on the development of AMS.

Supplier Relationship Management (SRM)
SRM practices create common frame of reference to enable effective communication between enterprises.In agile environments, relationships and communication between suppliers should be flexible and responsive 58 .Relationships with suppliers in agile manufacturing are considered in 26, 59 .Accordingly, we theorize the following hypothesis: H15 0 : SRM implementation has a significant, positive effect on the development of AMS.H15 1 : SRM implementation has no effect on the development of AMS.

Customer Relationship Management (CRM)
Traditional ways of communication with customers include Internet, business to customer B2C, business to business B2B.The Internet offers several advantages such as reduction of ordering process cost, revenue flow increase because of credit cards payment, global access, and pricing flexibility.In-house inventory placement, inventory pooling, forward placement, vendor-managed inventories VMI, and continuous replenishment program CRP may be used to build an effective agile customer relationship model.Hence, the following hypothesis is proposed: H16 0 : CRM implementation has a significant, positive effect on the development of AMS.
H16 1 : CRM implementation has no effect on the development of AMS.

Structural Equation Model (SME) and Research Hypotheses
The conceptual relationship model between the eight-construct latent considered in this study is shown in Figure 1.The relationships between the various-model latent are defined and summarized in Table 4.The relationship model is constructed based on authors' experience.Therefore, this model investigates the important relationships between the eight considered agile areas and the impacts of their implementations on the development of AMS.
Eight different questionnaire drafts were developed.The preliminary questionnaires were pilot tested and reviewed by managers of several industrial companies, extensive  literature review, and group of graduate students.This process continues until all questions in the eight questionnaires are unambiguous, appropriate, and acceptable to respondents.Every questionnaire is concerned with the implementation of one impact area.It consists of five-point Likert scale anchored at 1 "Poor", 2 "Fair", 3 "Good", 4 "Very good", and 5 "Excellent".

Data Collection and Analysis
Jordanian companies listed in Jordan chamber of commerce were screened according to whether they have a potential of implementing lean tools or not.Consequently, questionnaire packets were distributed to 500 services and manufacturing companies.456 companies have responded to the questionnaire packets.Data were collected through production managers, quality engineers, consultants, and owners.Cronbach's alpha α is a tool that measures and tests consistency validity and scale reliability.As shown in Table 5, Cronbach's alpha value of the whole AMS equals to 0.830 and the AGility index is 60.1%.AG is used to measure the overall implementation level of agile tactics in the studied sample.The results of reliability test indicate that both internal consistency and overall model reliability are high.Mean μ, variance σ 2 , area-tactic correlations, model-tactic correlations, tactic agility indexarea correlations, and area agility index are evaluated and summarized in Table 6.Agile tactics with no significant correlations at the 0.05 or less level 2 tailed are identified.The results of tactics-tactics correlation test are summarized in Table 7.
Interrelations between production areas are computed and investigated using correlation coefficients see Table 8 .It is observed that the correlation is significant at the

H16
H16 CRM implementation has a significant, positive effect on the development of AMS.

H16
CRM implementation has no effect on the development of AMS.

H15
H15 SRM implementation has a significant, positive effect on the development of AMS.

H15
SRM implementation has no effect on the development of AMS.

H14
H14 PPD implementation has a significant, positive effect on the development of AMS.

H14
PPD implementation has no effect on the development of AMS.

H13
H13 SFM implementation has a significant, positive effect on the development of AMS.

H13
SFM implementation has no effect on the development of AMS.

H12
H12 PPC implementation has a significant, positive effect on the development of SFM.

H12
PPC implementation has no effect on the development of SFM.

H11
H11 PPC implementation has a significant, positive effect on the development of AMS.

H11
PPC implementation has no effect on the development of AMS.MET implementation has no effect on the development of AMS.
and H16 1 are calculated.The calculated P-values are less than 0.05, which indicates that the proposed null hypotheses are true.The t-values fall within the 95% of the t-distribution −1.96 −t 0.025, 458 < t-Value < t 0.025, 458 1.96 .These results provide evidence that the alternative hypotheses are rejected.Influential dependencies see Table 10 between production areas are found, and hence multicollinearity is achieved.For the twotailed one-way ANOVA test at the 0.05 level, the f 0 -value as shown in Table 10 exceeds f 0.025,v1,v2 .This proves that all the considered agile tactics have a positive effect on AMS.The f 0 -value of the F-test obtained from the two-tailed one-way ANOVA analysis is less than 0.001.

Discussion of Results
This paper investigates the causal relationship model among implementation of thirty-six different agile tactics.These tactics are categorized into eight impact areas manufacturing equipment and technology MET, processes technology and know-how PTK, quality and productivity improvement and measures QPIM, production planning and control PPC, Shop Floor Management SFM, product design and development PDD, supplier relationship management SRM, and customer relationship management CRM .Analysis of data is carried out using AMOS 19 and IBM SPSS 20 for Windows.The obtained results show strongly that the model is valid.The AMOS 19 software is used to test the model fit for each impact area.The results show that the model fit is good.All items loaded significantly on their corresponding constructs at the 0.05 level.This demonstrates a good model fit.The fit statistics indicate that the hypothesized structural model achieves an acceptable fit such that no further interpretation is required.The testing of the entire hypotheses shows that all impact areas have positive effect on AMS.It was found out that the overall assumed agility index is about 60%, the average agility index of impact areas is about 60%, and the average agility index of agile tactics is about 60%.The correlation analyses show that all model constructs have a positive correlation with overall AMS model.
Estimates of the relations in the AMS are investigated and summarized as shown in Figure 2. The results of this research may be influenced by the person who fills the questionnaires.This may lead to errors due to the personal reliability and trustworthiness.

Conclusion
The implementation of agile manufacturing principles and tools in Jordanian firms is investigated.Different agile practices that are adopted by the considered firms to manage   their AMS systems are identified based on empirical basis.This paper concludes that the existence of 36 different agile approaches can be adopted by the different firms to enhance their competitiveness.These approaches categorized into eight impact areas, namely, MET, PTK, QPIM, PPC, SFM, PDD, SRM, and CRM.The primary contribution of this paper is successfully analyzing the causal relationship of implementation level of agile production areas and their effect on the AMS using SME methodology.The results ensure that SEM is the correct method for investigating the relationship model between the eight-constructs considered in this study.IBM SPSS 20 and AMOS 19.0.0 software enable SEM to provide a clear and complete specification of the AMS and its constructs.The results of this study show that the studied agile tactics have significant relationship and are affected positively by the AMS.The implementation of each agile tactic contributes significantly to the performance of AMS.The approach presented in this study can be used to facilitate the implementation of agile practices in industries and measure correlation between them.It may be worthwhile to focus future research on modeling the implementation of lean production practices, such as kanban, just in time JIT , pull production control strategy, and so forth, 60 and to compare and link the expected results with those concluded here.

H6 1 :
PTK implementation has no effect on the development of AMS.

H7 0 :
PTK implementation has a significant, positive effect on the development of QPIM.

H7 1 :
PTK implementation has no effect on the development of QPIM.

H8 1 :
QPIM implementation has no effect on the development of AMS.

H9 0 :
QPIM implementation has a significant, positive effect on the development of SRM.

H9 1 :
QPIM implementation has no effect on the development of SRM.

H10 0 :
QPIM implementation has a significant, positive effect on the development of CRM.

H10 1 :
QPIM implementation has no effect on the development of CRM.

H11 1 :
PPC implementation has no effect on the development of AMS.H12 0 : PPC implementation has a significant, positive effect on the development of SFM.

H12 1 :
PPC implementation has no effect on the development of SFM.

Figure 1 :
Figure 1: The proposed conceptual model and research hypotheses.

Table 8 :
Correlation matrix and fit indices for each production area.Correlation is significant at the 0.01 level 2 tailed .* Correlation is significant at the 0.05 level 2 tailed .

Figure 2 :
Figure 2: Estimates of the relations between models constructs.

Table 1 :
Mappings of AM tactics to eight impact areas.

Table 2 :
Comparison of agile manufacturing and lean production.

Table 3 :
Constructed latent of agile manufacturing system.
45.45.Software architecture for control of an agile manufacturing work cell is developed byKim et al. 46.Jacobs et al. provided a strong empirical evidence of the advantages of increasing the modularity of products in the firm's portfolio 47 .Chick et al. provided a descriptive model of the machining system selection process that is focused on capital intensive 48 .Swafford et al. found that information technology integration enables firms to utilize their flexibility 49 .Jacobs et al. studied the product and process modularity's effects on manufacturing agility and firm growth performance 47 .Forsythe summarized human factors contributions to the development of agile business practices and design of enabling technologies.The author also discussed human factors related to the communications and information infrastructure essential to organization to become agile 50 .Chunxia and Shensheng proposed a web-based agile architecture of supply chain management system 51 .Moore et al. proposed virtual manufacturing approach for designing, programming, testing, verifying, and deploying control systems for agile modular manufacturing machinery 52 .Based on previous discussion, we suggest the following hypothesis:

Table 4 :
Summary of relationships between various-model latent.
H10QPIM implementation has a significant, positive effect on the development of CRM.H10QPIM implementation has no effect on the development of CRM.

Table 5 :
Consistency and reliability of the model.

Table 8
. A good model fit is found.All items loading on their corresponding production area are high and significant at the 0.05 or less level 2 tailed .Significance level at 0.05 is recommended.

Table 6 :
Some statistics of model and correlation coefficients.

Table 9
summarizes the results of the hypothesis testing.The P values of the alterative hypotheses H1