group 6 - relation to jit, operational performance and firm performance

Journal Identification = OPEMAN Article Identification = 701 Date: March 15, 2011 Time: 2:7 pm

Ap

Ra

b

c

d

a

ARRAA

KAJOS

1

ibJplidcb

(tabm

((

0d

Journal of Operations Management 29 (2011) 343–355

Contents lists available at ScienceDirect

Journal of Operations Management

journa l homepage: www.e lsev ier .com/ locate / jom

gile manufacturing: Relation to JIT, operational performance and firmerformance

. Anthony Inmana,∗, R. Samuel Saleb, Kenneth W. Green Jr. c,1, Dwayne Whittend,2

College of Business, Louisiana Tech University, Box 10318, Ruston, LA 71272, United StatesDepartment of Management and Marketing, PO Box 10025, Lamar University, Beaumont, TX 77710, United StatesDepartment of Management, Marketing, and MIS, College of Business, Southern Arkansas University, P.O. Box 9410, Magnolia, AR 71754, United StatesTexas A&M University - Mays Business School, Information and Operations Management Department, Mailstop 4217, College Station, TX 77843, United States

r t i c l e i n f o

rticle history:eceived 7 January 2007eceived in revised form 1 June 2010ccepted 5 June 2010vailable online 18 June 2010

a b s t r a c t

A structural model incorporating agile manufacturing as the focal construct is theorized and tested.The model includes the primary components of JIT (JIT-purchasing and JIT-production) as antecedentsand operational performance and firm performance as consequences to agile manufacturing. Using datacollected from production and operations managers working for large U.S. manufacturers, the model is

eywords:gile manufacturing

IT systemsrganizational performancetructural equation modeling

assessed following a structural equation modeling methodology. The results indicate that JIT-purchasinghas a direct positive relationship with agile manufacturing while the positive relationship between JIT-production and agile manufacturing is mediated by JIT-purchasing. The results also indicate that agilemanufacturing has a direct positive relationship with the operational performance of the firm, that theoperational performance of the firm has a direct positive relationship with the marketing performanceof the firm, and that the positive relationship between the operational performance of the firm and the

the fi
financial performance of
. Introduction

Competitive pressures force manufacturers to continuouslymprove the provision of products and associated services desiredy customers. Manufacturers have adopted lean practices such asIT and TQM to reduce costs and improve quality. As many com-etitors adopted these practices, some competitive advantage was

ost. Many manufacturers now have begun adopting practices thatncrease their ability to rapidly respond to changes in customeremand. For these, superior responsiveness has become a key toompetitive advantage. In short, many manufacturing firms areecoming relatively more agile.

We propose that an element of lean manufacturing, Just-in-TimeJIT), is related to agile manufacturing. Specifically, we propose that
he primary elements of JIT, i.e., JIT-production and JIT-purchasing,re related to agility. Further we investigate the relationshipetween manufacturing agility and operational and firm perfor-ance.
∗ Corresponding author. Tel.: +1 318 257 3568; fax: +1 318 257 4253.E-mail addresses: [email protected] (R.A. Inman), [email protected]

R.S. Sale), [email protected] (K.W. Green Jr.), [email protected]. Whitten).

1 Tel.: +1 870 235 4317 (O).2 Tel.: +1 979 845 2919 (O).

272-6963/$ – see front matter © 2010 Elsevier B.V. All rights reserved.oi:10.1016/j.jom.2010.06.001

rm is mediated by the marketing performance of the firm.© 2010 Elsevier B.V. All rights reserved.

We conducted a national survey of production and operationsmanagers working for large U.S. manufacturing concerns to col-lect data necessary to assess the model using a structural equationmethodology. A review of the literature and discussion of the studyhypotheses follow in the next section. A discussion of the specificmethodology employed is followed by a description of the results ofthe scale assessment and the structural equation modeling results.Finally, a conclusions section, which incorporates discussions of thecontributions of the study, limitations of the study, suggestions forfuture related research, and implications for practicing managers,is provided.

2. Literature review and hypotheses

Shah and Ward (2003) identify JIT as one of four “bundles” thatmake up lean manufacturing. Given that JIT is an element of leanmanufacturing, discussion of the literature relating lean manufac-turing to agile manufacturing is relevant even though the currentstudy focuses on the relationship between JIT and agile manufactur-ing. Hence, the following section provides a review of the literaturefor both the JIT/agile relationship and the lean/agile relationship.

2.1. JIT and agile manufacturing

Specific to our research is the relationship between agilemanufacturing and the Just-in-Time (JIT) manufacturing strategy.

dx.doi.org/10.1016/j.jom.2010.06.001

http://www.sciencedirect.com/science/journal/02726963

http://www.elsevier.com/locate/jom

mailto:[email protected]




dx.doi.org/10.1016/j.jom.2010.06.001


3 tions

Cbsmdrf(mttpiQaamdapitr

pesteleieTdmibrcq

eWtioopoJe

2

mopwKpawws

44 R.A. Inman et al. / Journal of Opera

ountless research regarding JIT and its individual elements haseen generated in the last three decades. Claycomb et al. (1999b)tate that “in its ideal form, JIT integrates the entire supply chain’sarketing, distribution, customer service, purchasing, and pro-

uction functions into one controlled process.” In an early workegarding JIT implementation, Mehra and Inman (1992) identifiedour elements of JIT: JIT-production strategy, JIT vendor strategypurchasing), JIT education strategy and management commit-

ent. Only JIT-production and JIT vendor strategies were foundo have a significant impact on JIT implementation success. Sincehat time a number of published articles have at least partially sup-orted these findings. In more recent work Shah and Ward (2003)

dentify four “bundles” of lean production: Just-In-time (JIT), Totaluality Management (TQM), Total Preventive Maintenance (TPM)nd Human Resource Management (HRM). In a 2007 paper Shahnd Ward propose and test 10 dimensions that can be used toeasure these four “bundles” of lean production. Six of the 10

imensions are elements of JIT with three pertaining to supplierspects of JIT (purchasing) and three related to aspects of JIT-roduction. Therefore, while a number of JIT elements have been

dentified, two, JIT-production and JIT-purchasing, seem to garnerhe most support for their criticality to organization success. As aesult, we limit our work here to these two primary elements of JIT.

We define JIT as a comprehensive strategy that combines therimary tactical elements of JIT-production and JIT-purchasing, toliminate waste and optimally utilize resources throughout theupply chain (Claycomb et al., 1999b). JIT-production focuses onhe identification and elimination of all forms of waste, includingxcess inventories, material movements, production steps, scraposses, rejects and rework, within the production function (Wisnert al., 2005; Brox and Fader, 2002). JIT-purchasing is operational-zed by Freeland (1991) as a “set of techniques and concepts forliminating waste and inefficiency in the purchasing process.”echniques and concepts associated with JIT-purchasing includeaily delivery of small lot sizes from nearby vendors, shared infor-ation, supplier education, reduced inspection and early supplier

nvolvement in product/process design. The techniques utilizedy JIT-production and JIT-purchasing allow firms to translate theesulting capabilities into a JIT strategy that provides organizationalapabilities to deliver near zero defect quality, near zero varianceuantity and precise on-time delivery (Green and Inman, 2005).

The key word applicable to the definition of both primarylements of JIT is “waste.” This is consistent with Shah andard’s (2007) definition of lean production as an integrated socio-

echnical system with the main objective of reducing or eliminatingnternal, customer, and supplier waste. Since JIT is a subset (bundle)f lean, we narrow our definition to the following: JIT is that subsetf lean associated primarily with the elimination of waste throughlanning, scheduling and sequencing of operations. This definitionf JIT subsumes both primary elements of JIT, JIT-purchasing andIT-production, as elements of itself that are distinguishable fromach other by where they occur in the system or supply chain.

.2. Lean manufacturing and agile manufacturing

There has been a tendency to view the development of leananufacturing and agile manufacturing either in a progression

r in isolation (Gunasekaran, 1999a). From an isolation stand-oint, Harrison (1997) notes that companies with a lean mindsetould find the agile manufacturing concept difficult to follow.rishnamurthy and Yauch (2007) state that there are “three general
ositions with respect to lean and agile: those who believe that theyre mutually exclusive or distinct concepts that cannot co-exist, thoseho believe that they are mutually supportive strategies, and thoseho believe that leanness must be a precursor to agility.” Table 1
ummarizes the literature supporting each of the three views.

Management 29 (2011) 343–355

2.2.1. Lean and agile as mutually exclusive conceptsEarly concerns that the two concepts cannot co-exist were

expressed by Richards (1996), who noted that some agile propo-nents claimed that flexibility would suffer under lean productionand from Harrison (1997) who expressed doubts that lean andagile were compatible while emphasizing that agile implied moreresources, not fewer. More recently, Goldsby et al. (2006) note thatlean and agile are often pitted as opposing paradigms.

Agility has been recognized as a manufacturing strategy consist-ing of manufacturing tasks and choices (Gunasekaran et al., 2008).The word “choices” implies that tradeoffs are necessary betweenlean and agile (Harrison, 1997) or that they cannot completely co-exist. While both strategies address the same competitive priorities(cost, quality, service, flexibility), they each emphasize differentelements (Narasimhan et al., 2006) such that clear dividing linescan be drawn between the two (Gunasekaran et al., 2008). Somewould state that lean manufacturing subordinates responsiveness(service) to efficiency and productivity (cost) (Vazquez-Bustelo etal., 2007) while agile manufacturing focuses on speed and flexibil-ity and not cost (Gunasekaran et al., 2008). One may consider lean’smarket winner as cost (Christopher and Towill, 2001) and agile’smarket winners as speed, flexibility and responsiveness to changes(Zhang and Sharifi, 2007), i.e., service level (Mason-Jones et al.,2000). This is consistent with Narasimhan’s et al. (2006) empiricalstudy that found agile plants to meet/exceed lean plants and otherplants in all measured performance dimensions with the exceptionof cost efficiency. Hence, tradeoffs that would prevent lean/agileco-existence can be easily envisioned. Larger lot sizes and higherinventory levels could be necessary to maintain the higher ser-vice level required by agile firms while smaller lot sizes and lowerinventory levels could be required by cost-efficient lean firms.

It should be noted that there is a stream of thought thatadvocates the simultaneous use of lean manufacturing and agilemanufacturing. Termed “leagile,” proponents believe that man-ufacturing systems can consist of both lean and agile, actingtogether to “exploit market opportunities in a cost-efficient man-ner” (Krishnamurthy and Yauch, 2007). However, this appears tobe appropriate only for supply chains, not individual manufactur-ing firms unless the firm is a multi-unit enterprise that functionsas a supply chain. Leagile models created thus far contain a decou-pling point that separates the lean and agile portions of the system(Krishnamurthy and Yauch, 2007) with the lean portion on theupstream side of the point and the agile portion of the system on thedownstream side (Mason-Jones et al., 2000). Krishnamurthy andYauch (2007) state that this decoupling point ensures that lean andagile do not co-exist, lending credence to the idea that the two aremutually exclusive within a single manufacturing entity, althoughboth may exist within a supply chain.

From the literature, one can glean that both lean and agile haveobtained desired results in isolation and that neither is better norworse than the other (Naylor et al., 1999). This would imply thateither could be used successfully depending upon the individualfirm’s environment. Specifically, lean manufacturing is appropri-ate when market conditions are basically stable, demand is smoothand standard products are produced and agile manufacturing isappropriate when the environment is more turbulent and moreproduct variety is present (Vazquez-Bustelo et al., 2007; Nayloret al., 1999). The degree of turbulence in the environment deter-mines the degree of agility needed (Vazquez-Bustelo et al., 2007;Sharifi and Zhang, 2001; Zhang and Sharifi, 2000). Though notstated within the literature, the same could hold true for lean. The
degree of stability dictates the degree of leanness required to effec-tively compete. Consistent with the above, Goldsby et al. (2006)found, via simulation, that a lean strategy resulted in the lowestcost/highest service when demand was smooth and predicted witha high degree of accuracy coupled with low-value finished goods


R.A. Inman et al. / Journal of Operations Management 29 (2011) 343–355 345

Table 1Three views of the relationship between lean (JIT) and agile manufacturing.

Lean and agile as mutually exclusive conceptsHarrison (1997) Expressed doubts that lean and agile were compatible while emphasizing that agile implied more

resources, not fewerGoldsby et al. (2006) Note that lean and agile are often pitted as opposing paradigmsNarasimhan et al. (2006) They each emphasize different elementsGunasekaran et al. (2008) Clear dividing lines can be drawn between the two; agile manufacturing focuses on speed and

flexibility and not costVazquez-Bustelo et al. (2007) Lean manufacturing subordinates responsiveness (service) to efficiency and productivity (cost)Christopher and Towill (2001) Lean’s market winner is costZhang and Sharifi (2007); Mason-Jones et al. (2000) Agile’s market winners are speed, flexibility and responsiveness to changes, i.e., service levelVazquez-Bustelo et al. (2007); Naylor et al. (1999) Lean manufacturing is appropriate when market conditions are basically stable, demand is smooth

and standard products are produced and agile manufacturing is appropriate when theenvironment is more turbulent and more product variety is present

Lean and agile as mutually supportive conceptsKatayama and Bennett (1999) Leanness is an overarching concept that is compatible with any production systemKatayama and Bennett (1999); Krishnamurthy and Yauch

(2007)Mutually supportive concepts

Krishnamurthy and Yauch (2007) Results in benefits not accessible when the concepts are used in isolationKidd (1994) Compatible conceptsNaylor et al. (1999) Complementary conceptsGunasekaran et al. (2008); Ramesh and Devadasan (2007);

Goldsby et al. (2006): McCullen and Towill (2001)Elements cited as necessary for agile performance include elements of lean manufacturing,specifically Just-in-Time manufacturing

Lean as antecedent to agilityNarasimhan et al. (2006) The predominant view in the literature is that “lean manufacturing is a performance/practice state

that is antecedent to agile manufacturing”Jin-Hai et al. (2003); Hormozi (2001) Agile is the latest step in the evolution from mass production, to Just-in-Time to lean to agileGoldman and Nagel (1993) Agile manufacturing assimilates the full range of flexible production technologies, along with the

lessons learned from TQM, JIT, and lean productionGunasekaran et al. (2008); Vazquez-Bustelo et al. (2007);

Sharifi and Zhang (2001); Zhang and Sharifi (2000)Agile manufacturing can be achieved by utilizing and integrating elements of existing systems andmethods that are already developed and in use

Sarkis (2001) Agile manufacturing = flexible manufacturing system + lean manufacturingMcCullen and Towill (2001) Agile manufacturing can subsume the paradigm of lean production

acturiing

a(to

2

cBa1Kiieocpt2sefc

2

aaeZt

Gunasekaran et al. (2008); Hormozi (2001); Maskell(2001); Gunasekaran (1999b); Robertson and Jones(1999); Booth (1996)

Agile manufmanufactur

nd low carrying costs. Results of a study by Narasimhan et al.2006) indicated that lean performers had “made-to-stock” opera-ions while agile performers had a significantly greater proportionf “to-order” operations.

.2.2. Lean and agile as mutually supportive conceptsAlternately, leanness has been described as an overarching con-

ept that is compatible with any production system (Katayama andennett, 1999) and as such should be compatible (Krishnamurthynd Yauch, 2007; Kidd, 1994), complementary (Naylor et al.,999), and mutually supportive (Krishnamurthy and Yauch, 2007;atayama and Bennett, 1999) with agile manufacturing, result-

ng in benefits not accessible when the concepts are used insolation (Krishnamurthy and Yauch, 2007). Elements cited as nec-ssary for agile performance include: the ability to produce larger small batches with minimum setups (and setup time) and aross-trained flexible workforce (Goldsby et al., 2006); reducedrocess lead times and costs (Gunasekaran et al., 2008); rela-ionships with suppliers and JIT-production (McCullen and Towill,001); fully empowered employees, JIT-purchasing, and flexibleetups (Ramesh and Devadasan, 2007). Interestingly, these are alllements of lean manufacturing, specifically Just-in-Time manu-acturing. Based on the above logic, it seems that the two conceptsould indeed be mutually supportive.

.2.3. Lean as antecedent to agilityA number of researchers feel that agile manufacturing can be

chieved by utilizing and integrating elements of existing systemsnd methods that are already developed and in use (Gunasekarant al., 2008; Vazquez-Bustelo et al., 2007; Sharifi and Zhang, 2001;hang and Sharifi, 2000). More specifically, there are those that feelhat agile manufacturing is the next logical step or a natural devel-

ng is the next logical step or a natural development from the concept of lean

opment from the concept of lean manufacturing (Gunasekaranet al., 2008; Hormozi, 2001; Maskell, 2001; Gunasekaran, 1999b;Robertson and Jones, 1999; Booth, 1996). Sarkis (2001) offersthe formula: agile manufacturing = flexible manufacturing sys-tem + lean manufacturing. McCullen and Towill (2001) argue thatagile manufacturing can subsume the paradigm of lean produc-tion. Specific, to our research, Narasimhan et al. (2006) reportthat the predominant view in the literature is that “lean man-ufacturing is a performance/practice state that is antecedent toagile manufacturing,” with results of their study suggesting thatleanness is a precursor to agility. One may summarize this partof the literature review with Goldman and Nagel’s (1993) state-ment that agile manufacturing “assimilates the full range of flexibleproduction technologies, along with the lessons learned from Total-Quality-Management [an element of lean, Shah and Ward, 2003],Just-in-Time production [an element of lean, Shah and Ward, 2003]and lean production.”

Perusing the literature review also begs the question, “when islean manufacturing assimilated into the agile system?” Does it haveto be established before moving on to agile manufacturing, can alean system be established at the same time and as a part of an agilesystem, or does it really matter? A number of researchers state thatagile is the latest step in the evolution from mass production, toJust-in-Time to lean to agile (Jin-Hai et al., 2003; Hormozi, 2001). Ifthis is the case, then most agile firms probably adopted lean at somepoint and then later moved on to agile, making lean a precursor toagile. Simply stated by Narasimhan et al. (2006), “results indicate
that while the pursuit of agility might presume leanness, pursuit ofleanness might not presume agility.”
Since most of the evidence put forth by the “precursor” litera-ture would just as well justify a “mutually supportive” stance, wemake the assumption that if lean is antecedent to agile, as proposed,


3 tions

m((ini

2

tdtVJ(oleoaJeh

Hti

Hti

rss

iaa2


utual support will also be present. This leaves us with two options1) lean manufacturing (JIT) is antecedent to agile manufacturingwith mutual support assumed), i.e., higher levels of JIT will resultn higher levels of agility, or (2) they are distinct concepts that can-ot co-exist, that is, the increased effectiveness in one area results

n a decrease in effectiveness in the other.

.3. Hypotheses

As stated earlier, Shah and Ward (2003) identified JIT as one ofhe four bundles that make up lean manufacturing so the precedingiscussion involving lean manufacturing is assumed to apply alsoo the specific lean bundle, JIT. This assumption is supported byazquez-Bustelo et al. (2007) who state that experience suggests a

IT-production system is required for agility and Narasimhan et al.2006) who found that supplier management and JIT flow and lay-ut received a significantly higher emphasis in agile firms than inean firms. Also, as previously noted, many elements cited as nec-ssary for agile performance are previously established elementsf JIT. Hence, our research question becomes “Is JIT antecedent togile manufacturing (thus, the two are mutually supportive) or areIT and agile manufacturing two distinct concepts that cannot co-xist?” Using the two primary elements of JIT, we propose twoypotheses to define our research objective.

1. Higher levels of adoption of a JIT-purchasing strategy will leado higher levels of a firm’s manufacturing agility, i.e., JIT-purchasings antecedent to agility.

2. Higher levels of adoption of a JIT-production strategy will leado higher levels of a firm’s manufacturing agility, i.e., JIT-productions antecedent to agility.

Hypotheses 1 and 2 are conceptually pictured in Fig. 1. If testesults indicate a significant negative relationship between the JITtrategies and agile manufacturing, a mutually exclusive relation-hip between the two will be supported.

Manufacturers become more agile with the expectation ofmproving performance (Yusuf and Adeleye, 2002; Mason-Jones etl., 2000). Organizational performance encompasses both financialnd marketing performance at the firm level (Green and Inman,005; Green et al., 2004). Financial performance focuses on a firm’s

Fig. 1. Agile manufacturing m

Management 29 (2011) 343–355

return on investment, return on sales and profitability as comparedto its competition. The marketing performance component com-pares the firm’s sales volume, sales growth, and market share tothat of its competition.

Yusuf and Adeleye (2002) surveyed 109 manufacturers andfound a significant link between agility and business performance(sales turnover, market share, customer loyalty, performance rela-tive to competitors, and aggregate performance). Vazquez-Busteloet al. (2007) surveyed firms in Spain and found that agile manu-facturing positively impacted manufacturing strength which led toimproved operational, financial and market performance. Resultsof a survey by Narasimhan et al. (2006) revealed that agile plantsmet or exceeded lean and other plants in all measured performancedimensions except cost efficiency, giving agility the appearance ofa higher state of plant performance and capability.

We propose that firms adopting agile manufacturing practiceswill experience improved operational and firm performance. Thefollowing hypotheses were fashioned based upon this proposition:

H3. Higher levels of manufacturing agility will have a positiveimpact on a firm’s financial performance.

H4. Higher levels of manufacturing agility will have a positiveimpact on a firm’s marketing performance.

H5. Higher levels of manufacturing agility will have a positiveimpact on a firm’s operational performance.

Hypotheses 3–5 are also conceptually pictured in Fig. 1.

3. Methodology

A listing of 1350 plant and operations managers was extractedfrom the 2004 Manufacturer’s News, incorporated database ofU.S. manufacturers with more than 250 employees. Plant andoperations managers were targeted because of their particularknowledge related to the manufacturing processes within their
organizations. It was assumed that plant and operations managers,as a group, would be interested in participating in the survey andwould readily understand the survey items. The survey instrumentwas moderately long, filling the front and back of two legal-sizepages.
odel with hypotheses.


tions

ttcaatfvtw

lCoaaamti

3

Jpt

fldweaitsa

opmwhm(mw

fTiRcfsb5p

madiss

R.A. Inman et al. / Journal of Opera

Each of the manufacturers was mailed an initial request to par-icipate that included a cover letter, a “non-participating” form,he survey instrument, and a postage-paid return envelope. Theover letter requested participation and stated an assurance thatll responses would be anonymous. The “non-participating” formllowed plant managers who did not wish to participate in the studyo have their names and addresses removed from the database. Aollow-up mailing that included a revised cover letter, another sur-ey instrument, and return envelope was completed 2 weeks afterhe initial mailing. This second mailing did not include managersho filled out the “non-participating” form.

A descriptive profile of respondents was prepared, and early andate responders were compared to assess for non-response bias.onfirmatory factor analysis was used to assess the dimensionalityf the study scales, and all scales were further assessed for reliabilitynd validity. Summary values were computed for each study vari-ble. Descriptive statistics for each of the variables were computednd a correlation matrix prepared. A structural equation modelingethodology was used to determine how well the agile manufac-

uring model fit the data and to identify support for each of thencorporated hypotheses.

.1. Measurement of constructs

The theorized model incorporates constructs related toIT-production, JIT-purchasing, agile manufacturing, operationalerformance, and firm performance. The scales selected to measurehe constructs are displayed in Appendix A.

Agile manufacturers must exhibit capabilities of responsiveness,exibility, and quickness in responding to changes in customeremand (Sharifi and Zhang, 2001). The agile manufacturing scaleas developed based on a prioritized listing of 20 capabilities nec-

ssary for organizations to achieve agility developed by Sharifind Zhang (2001). A scale item was fashioned for each of Shar-fi and Zhang’s top 10 items. Respondents were asked to indicateheir degree of agreement with each statement. Seven-point Likertcales were used with “strongly disagree” and “strongly agree” asnchors.

JIT-production and JIT-purchasing focus on the eliminationf waste and optimal utilization of resources in production andurchasing processes. JIT-production was measured using theulti-item scale developed by Brox and Fader (2002). Respondentsere asked to indicate which of 13 JIT-production related practicesad been implemented by their organizations. JIT-purchasing waseasured with the 7-item scale developed by Germain and Dröge

1997). Respondents were asked to indicate their degree of agree-ent with each statement. Seven-point Likert scales were usedith “strongly disagree” and “strongly agree” as anchors.

Operational performance was measured using a 13-item “per-ormance metrics” scale developed by Bowersox et al. (2000).he items incorporate customer service, cost management, qual-ty, productivity and asset management performance metrics.espondents were asked to rate their organization’s performanceompared to that of their competitors on the operational per-ormance metrics. The items were measured using 7-point Likertcales anchored with “much worse than competition” and “muchetter than competition.” Although Bowersox et al. (2000) used-point scales, the 7-point scales were adopted for consistencyurposes.

The scales for measuring the financial and marketing perfor-ance of the firm were previously used by Green and Inman (2005)

nd Green et al. (2004). The financial performance items were takenirectly from Claycomb et al. (1999a). The marketing performance

tems were developed by Green and Inman (2005) based on mea-ures of marketing performance (sales volume, market share andales growth) identified by Kohli and Jaworski (1990). The items

Management 29 (2011) 343–355 347

in these scales were measured with 7-point Likert scales anchoredwith “strongly disagree” and “strongly agree.”

4. Results

4.1. Survey effectiveness

A total of 1350 packets were mailed of which 18 were returneddue to incorrect addresses. Further, 121 “non-participating” formswere returned. Ninety-six manufacturers responded with com-pleted instruments for a response rate of 7.9%. This response rateis low but not atypical for industrial research. Other publishedworks in similar circumstances yielded response rates as low as7.5% (Nahm et al., 2003a,b), 6.7% (Tan et al., 2002), and 6.3% (Dwyerand Welsh, 1985). While Patterson et al. (2004) did not specificallyidentify their response rate, they found it necessary to survey threedifferent databases (one of higher-level managers and two of logis-tics managers) to gather only 107 responses. While manufacturingmanagers are the prime source for supply chain managementrelated data, they are often under severe time and resource con-straints making it difficult to achieve high response rates to surveys.

Lambert and Harrington (1990, p. 21) describe a commonapproach to assessment as comparing the first and second wavesof responses and assuming that “non-response bias is nonexis-tent if no differences exist on the survey variables.” Following thiscommon approach, respondents were categorized as respondingto either the initial or follow-up requests sent approximately 2weeks later. Those responding to the initial requests were classifiedas early responders; those responding to the follow-up requestswere classified as late responders. Fifty-four percent (52) of therespondents were categorized as early respondents and 46% (44)were categorized as late respondents. A comparison of the meansof the descriptive variables and the scale items for the two groupswas conducted using one-way ANOVA. The comparisons resultedin statistically non-significant differences at the .01 level. Becausenon-respondents have been found to descriptively resemble laterespondents (Armstrong and Overton, 1977), this finding of gen-eral equality between early and late respondents indicates thatnon-response bias has not negatively impacted the assembled dataset.

When data for the independent and dependent variables arecollected from single informants, common method bias may leadto inflated estimates of the relationships between the variables(Podsakoff and Organ, 1986). As Podsakoff and Organ (1986) rec-ommended, Harman’s one-factor test was used post hoc to examinethe extent of the potential bias. As prescribed by Harman’s test, allvariables were entered into a principal components factor analy-sis. According to Podsakoff and Organ (1986), substantial commonmethod variance is signaled by the emergence of either a singlefactor or one “general” factor that explains a majority of the totalvariance. Results of the factor analysis revealed seven factors witheigenvalues greater than one, which combined to account for 70%of the total variance. While the first factor accounted for 30% of thetotal variance, it did not account for a majority of the variance.Based upon these results of Harman’s one-factor test, problemsassociated with common method bias are not considered signifi-cant (Podsakoff and Organ, 1986).

4.2. Sample description

All of the respondents indicated that they worked for manu-facturing organizations. Seventy-two percent of the respondentsidentified themselves specifically as plant or operations managers.The remaining 28% held management positions related to manu-facturing, purchasing and distribution. Respondents averaged 5.5


348 R.A. Inman et al. / Journal of Operations Management 29 (2011) 343–355

Table 2Scale assessment.

Reliability coefficients

Scale GFI RMSEA NNFI CFI Alpha Construct-reliability Variance-extracted NFI

Agile manufacturing .937 .079 .973 .982 .85 .89 .51 .956JIT-production .982 .087 .974 .991 .78 .82 .54 .979JIT-purchasing .971 .029 .991 .995 .79 .53 .42 .966Operational performance * * * * .80 .80 .58 *

0

yieSwoufI

4

dv(2iT

papcattv

entbi7rf

v((i

riaApaatve

Marketing performance * * * *

Financial performance .996 .000 1.000 1.00

* Values not available for scales containing only 3 items.

ears in their current positions. Mean sales revenues for the firmsncluded in the sample were $7.7 billion, and the mean number ofmployees per firm was 21,211. Seventeen specific manufacturingIC codes were identified. The most frequently identified SIC codesere: 34-fabricated metal products at 15.6%, 36-electronic and

ther electrical equipment at 7.3%, and 20-food and kindred prod-cts at 7.3%. Respondents represented 30 different states. The mostrequently identified states were Ohio (10.4%), Michigan (9.4%), andllinois (7.3%).

.3. Scale assessment process

Quality measurement scales must exhibit content validity, uni-imensionality, reliability, discriminant validity, and convergentalidity. Since all scales were taken directly from prior researchSharifi and Zhang, 2001; Claycomb et al., 1999a,b; Brox and Fader,002; Green and Inman, 2005; Bowersox et al., 2000), content valid-

ty is assumed. Survey results used to assess all scales are found inable 2.

With the exception of operational performance and JIT-roduction, all scales were treated as first-order factors (Garvernd Mentzer, 1999). Bowersox et al. (2000) described operationalerformance as comprised of five distinct factors: customer service,ost management, quality, productivity and asset management. Tossess unidimensionality, operational performance was, therefore,reated as a second-order construct. Values for each of the five fac-ors were calculated by averaging across factor items, and the factoralues were used in the unidimensionality assessment.

Because responses to the JIT-production scale items were cat-gorical (either “implemented” or “not implemented”), it wasecessary to compute four composite measures by summing acrosshe individual items in a manner similar to that recommendedy Garver and Mentzer (1999). The original scale includes 17

tems. KR20 reliability analysis indicated that the removal of item(Preventive Maintenance Programs) would improve the overall

eliability of the scale. The remaining 16 items were segmented intoour groups to facilitate computation of the composites.

Unidimensionality is indicated by goodness-of-fit index (GFI)alues greater than .90 (Ahire et al., 1996), non-normed-fit indexNNFI) and comparative-fit index (CFI) values greater than .90Garver and Mentzer, 1999), and root mean square error of approx-mation (RMSEA) below .08 (Garver and Mentzer, 1999).

In order to achieve unidimensionality, it was necessary toemove items 6, 8 and 10 from the agile manufacturing scale andtem 7 from the JIT-purchasing scale, and the cost management andsset management factors from the operational performance scale.fter re-specification the agile manufacturing, JIT-production, JIT-urchasing, and financial performance scales all met the GFI, NNFI,
nd CFI minimums indicating unidimensionality. Because the oper-tional performance and financial performance scales contain onlyhree items, it is not possible to compute GFI, NNFI, CFI, and RMSEAalues. Principal components analysis, however, indicated thatach scale measured only one dimension. The parameter estimates
.91 .95 .86 *

.92 .93 .77 .998

for all scales were significant and greater than .60 also indicat-ing unidimensionality. The RMSEA values for the JIT-purchasing,agile manufacturing, and financial performance were below therecommended .08 level. The RMSEA for JIT-production only slightlyexceeded the .08 level at .087.

Garver and Mentzer (1999) recommend computing Cronbach’scoefficient alpha and the SEM construct-reliability and variance-extracted measures to assess scale reliability. They indicate thatalpha and construct-reliability values greater than or equal to .70and a variance-extracted measure of .50 or greater indicate suf-ficient reliability. Two of the three reliability coefficients for theJIT-purchasing scale exceed the recommended minimums (at .42,the variance-extracted measure was slightly below the desired .5).All other scales exceed the minimum reliability requirements onall three measures.

Ahire et al. (1996) recommend assessing convergent validityusing the normed-fit index (NFI) coefficient with values greaterthan .9 indicating strong validity. Garver and Mentzer (1999) rec-ommend reviewing the magnitude of the parameter estimates forthe individual measurement items to assess convergent validity.A strong condition of validity is indicated when the estimates arestatistically significant and greater than or equal to .70. A weakcondition of validity is indicated when estimates are statisticallysignificant but have values less than .70.

While the NFI was not available for the operational perfor-mance and marketing performance scales, significant parameterestimates greater than .70 indicate convergent validity for both.An NFI exceeding .95 and statistical significance of the parame-ter estimates indicates sufficient convergent validity for all otherscales.

Gerbing and Anderson (1988) recommend that scales be testedfor discriminant validity using a chi-square difference test for eachpair of scales under consideration. A statistically significant dif-ference in chi-squares indicates discriminant validity (Garver andMentzer, 1999; Ahire et al., 1996; Gerbing and Anderson, 1988).The Chi-square difference tests for pairings of each scale with otherstudy scales returned significant differences at the .01 level, indi-cating discriminant validity for all scales.

4.4. Measurement model

Fig. 2 displays the measurement model that incorporates thescales described and assessed in the preceding section. As Koufteros(1999) recommends, the scales are further assessed within the con-text of the full measurement model using a confirmatory factoranalysis methodology. The measurement model fits the data rela-tively well with a relative chi-square value of .95, an RMSEA valueof 0.00, a CFI value of .99, and an NNFI value of 0.99. A review of the
standardized residual matrix identified only four pairs with abso-lute values greater than 2.58 (OPA and MP5, JITPRB and FP4, JITPRBand MP6, and JITPRB and MP7). Looking at the individual items, wechose not to re-specify based on the importance of each item to theaffected scales.



tive c

4

aotasscp

m

Fig. 2. Measurement model with standardized estimates and (t-values). Rela

.5. Structural equation modeling results

Summary values for the study variables were computed by aver-ging across the items in the re-specified scales with the exceptionf the JIT-production scale for which items were summed. Descrip-ive statistics and the correlation matrix for the summary variablesre presented in Table 3. Correlation coefficients are positive andignificant at the .05 level for all of the hypothesized relation-
hips in the agile manufacturing model with the exception of theoefficients for JIT-purchasing and financial performance, and JIT-roduction and operational performance.
Fig. 1 depicts the theorized agile manufacturing performanceodel as structurally assessed. Fig. 3 illustrates the model with the

hi-square = .95; Chi-square P-value = 0.72; RMSEA = 0.00; CFI = 0.99; NNFI = 0.99.

structural equation modeling results specified in the LISREL 8.7 out-put. Results relating to fit of the model generally support a claimof good fit. The relative chi-square (chi-square/degrees of freedom)value of 1.08 is less than the 3.00 maximum recommended by Kline(1998) and the root mean square error of approximation (.03) isbelow the recommended maximum of .08 (Schumacker and Lomax,1996). The P-value associated with the chi-square is .17, abovethe recommended minimum of.05 (Byrne, 1998). Results associ-
ated with the fit indices are somewhat mixed. The GFI (.79) andNFI (.88) are below the .90 level recommended by Byrne (1998).These indices are more heavily impacted by a relatively small sam-ple size and, as Byrne (1998) points out, the Comparative-Fit Index(CFI) and Incremental-Fit Index (IFI) are more appropriate when


350 R.A. Inman et al. / Journal of Operations Management 29 (2011) 343–355

Table 3Descriptive statistics and correlations.

Summary variable Mean Standard deviation

A. Descriptive statistics (n = 96)Agilemanufacturing(AM) 4.89 .92

JIT-purchasing (JITPU) 4.61 1.14JIT-production (JITPR) 9.42 3.79Operational performance (OP) 5.26 .69Financial performance (FP) 4.69 1.20Marketing performance (MP) 4.51 1.23

AM JITPU JITPR OP FP

B. Correlation matrix (n = 96)JITPU .484**

JITPR .322** .531**

OP .477** .316** .145FP .421** .134 .206* .321**

tr

pTsm(mio(aoa

pTre

FR

MP .377** .224*

* Correlation is significant at the 0.05 level (2-tailed).** Correlation is significant at the 0.01 level (2-tailed).

he sample size is small. The CFI (.98) and IFI (.98) both exceed theecommended .90 level (Byrne, 1998).

Of the five study hypotheses, only the relationship between JIT-roduction and agile manufacturing is not supported by the results.he JIT-purchasing to agile manufacturing link (H1) is positive andignificant with an estimate of .65 and t-value of 3.17. The esti-ate of −.10 for the link from JIT-production to agile manufacturing

H2) is non-significant with a t-value of −.56. The link from agileanufacturing to financial performance (H3) is positive and signif-

cant with a standardized estimate of .49 and an associated t-valuef 3.72. The agile manufacturing to marketing performance linkH4) is positive and significant with a standardized estimate of .46nd associated t-value of 3.67. Finally, the agile manufacturing toperational performance link (H5) is positive and significant withstandardized estimate of 0.58 and t-value of 4.13.

The lack of support for the hypothesized link between JIT-roduction and agile-manufacturing is surprising and troubling.his result, combined with the modification indices, led us toethink the model. This change in thought, coupled with Hairt al. (1998) recommendation for a competing models approach

ig. 3. Agile manufacturing hypothesized structural model with standardized estimateselative chi-square = 1.08; Chi-square P-value = 0.17; RMSEA = 0.03; CFI = 0.98; NNFI = 0.97

.251* .386** .640**

to structural equation modeling when alternative formulationsare suggested by underlying theory, prompted us to remove thepath between JIT-production and agile manufacturing, making JIT-production antecedent to JIT-purchasing, thereby, indicating anindirect (mediation), rather than direct, link to agile manufacturing.While material purchase obviously must occur before produc-tion, most purchasing is based on production plans that anticipatescheduling and sequencing activities.

Additionally, a review of the modification indices, resulting fromassessment of the theorized model, suggests that an additional pathfrom marketing performance to financial performance be added.Inclusion of the additional path is supported by the results of astudy by Green et al. (2006) which reported a positive relation-ship between marketing performance and financial performance.Vazquez-Bustelo et al. (2007) found that the adoption of agile man-
ufacturing positively impacts manufacturing strength thus leadingto improved business performance, hence operational performancewas treated as an antecedent to firm performance, i.e., marketingand financial, performance. This alternative model and associatedstructural equation modeling results are illustrated in Fig. 4.
(** significant at 0.01 level)..



F (** sigR 0.99.

pafptJaomti3pTiwCr

eadHnoA

5

asmsms

amlss

ig. 4. Agile manufacturing good-fit structural model with standardized estimateselative chi-square = .96; Chi square P-value = 0.70; RMSEA = 0.00; CFI = 0.99; NNFI =

The standardized estimate for the JIT-production to JIT-urchasing link is .69 with an associated t-value of 4.88 (significantt the .01 level). The link from JIT-purchasing to agile manu-acturing remains positive and significant. Rather than a directositive relationship between JIT-production and agile manufac-uring, based on this reformulation of the model, it appears thatIT-purchasing mediates the relationship between JIT-productionnd agile-manufacturing. The link from agile-manufacturing toperational performance remains positive and significant. The esti-ate for the marketing to financial performance link is 0.64 with a

-value of 4.96. Operational performance directly impacts market-ng performance with a standardized estimate of .44 and t-value of.49. Operational performance does not directly impact financialerformance, however, with an estimate of .11 and t-value of 1.01.he impact of operational performance on financial performances mediated by marketing performance. The overall fit improved

ith a relative chi-square = .96, a RMSEA of 0.00, a P-value of .70, aFI of .99 and an NNFI of .99. The GFI (.81) and NFI (.89), however,emained below the desired .90 level.

There was concern that environmental uncertainty may mod-rate the hypothesized relationship between agile manufacturingnd operational performance. Following the general methodologyescribed by Baron and Kenny (1986), moderation was assessed.owever, the results indicated that environmental uncertainty didot moderate the relationship between agile manufacturing andrganizational performance. Details of the analysis are found inppendix B.

. Discussion

A broad sample of large U.S. manufacturers provided data forssessing the agile manufacturing performance model. Althoughome re-specification was necessary, all study scales were deter-ined to be unidimensional, reliable, and valid. Results of the

tructural equation modeling analysis showed that the overallodel fit the data well and specifically support all but one of the

tudy hypotheses.The resulting support for the idea that JIT-purchasing is

ntecedent to agile manufacturing is not surprising. Within theanufacturing sector increased use of JIT-purchasing practices

ead to improved agile manufacturing capabilities. This partiallyupports the theoretical literature that purports that leanness,pecifically JIT implementation, is a “foundation” or a precur-

nificant at 0.01 level).

sor to agility (mutually supportive) and the empirical findings ofNarasimhan et al. (2006) that “when viewed from a performance[capability] perspective, leanness is a precursor to agility.” Sur-prisingly only one of the two primary elements of JIT was foundto support agility. The relationship between JIT-production andagile manufacturing was non-significant. This result would seemto indicate that, within the context of the model, JIT-purchasingalone, rather than in combination with JIT-production, explainsa significant portion of the variation in agile manufacturing. Thisis inconsistent with the belief that JIT flow and other productionrelated activities are precursors to agility. However, Narasimhanet al. (2006) note that other studies have shown that JIT flow isless significant than other elements. Our finding does not supportH2. The results for JIT-production do not support the notion thatJIT-production is antecedent to agile manufacturing nor does it sup-port the notion that the two are mutually exclusive. This may be inagreement with McCullen and Towill’s (2001) argument that agilemanufacturing can subsume the paradigm of lean production. Is theproduction aspect of JIT so much a part of agility that one may notdistinguish a difference between JIT-production and the produc-tion element within an agile manufacturing firm? Narasimhan et al.(2006) note that “agile does imply that many of the principles andtechniques of lean manufacturing are in place.” If the JIT-productionelement is already in place then increased supplier/customer inte-gration, in the form of high levels of JIT-purchasing, could showa far greater impact on agility than JIT-production alone. Thisis one possible explanation for the lack of support for Hypoth-esis 2 in the original model. The “move” from JIT to agile couldinvolve keeping the JIT-production element of lean constant butgreatly increasing the emphasis on JIT-purchasing. Although nosuch move is tested directly, the conceptual argument is consis-tent with the results of the alternate model in which JIT-productionis assessed as antecedent to JIT-purchasing. The idea that manu-facturing excellence may generate a certain level of performance,but that additional improvement requires the level of supplychain integration suggested by JIT-purchasing is consistent with“outward-facing” firms in Frohlich and Westbrook’s (2001) classi-fication of firms based on “arcs of integration.”

In addition it was found that organizations that become agilemanufacturers can expect improved operational and firm per-formance. This finding is consistent with Vazquez-Bustelo et al.(2007) who found that the adoption of agile manufacturing pos-itively impacts manufacturing strength thus leading to improved


3 tions

btmpbmao

5

prwnfmbtacaapstt

tsirswpt

taatem

5

atmicafmrbow

1. Orders are placed to suppliers and delivered on a daily basis.2. Our suppliers’ warehouses/factories are located nearby.


usiness performance and Narasimhan et al. (2006) who foundhat agile firms exceed lean and other firms on most perfor-

ance measures used. Interestingly, this is consistent with theroposed relationships among three of the four perspectives inalanced scorecard logic; customer perspective (marketing perfor-ance), internal business perspective (operational performance),

nd financial perspective (financial performance). In summary, weffer the following proposal:

In the manufacturing sector, JIT-purchasing combined with JIT-production enhances a firm’s manufacturing agility. Improvedmanufacturing agility leads to the improved operating per-formance of the firm, which in turn leads to the improvedmarketing and financial performance of the firm.

.1. Limitations of the study

While the objectives of the study were successfully accom-lished, limitations of the study should be noted. The response rateaised concerns of potential non-response bias. Although the twoaves of responses were compared and no evidence of bias wasoted, a more direct assessment of the potential bias utilizing data

rom a third wave and an intensive follow-up on non-respondentsay have strengthened the study. Because responses related to

oth the dependent and independent variables were collected fromhe same individual, the potential for common method bias waslso a concern. While subsequent testing for the bias relieved theoncern, collection of the strategy and performance data from sep-rate sources would also have strengthened the study. Also, sincell measures were at the “organization” level, not the individuallant level, data from multi-plant firms could dilute the data ifome plants were focused on lean and others on agility. However,his should weaken the results rather than artificially strengthenhem.

There is concern that the measurement scales conceptualizinghe JIT-related constructs were borrowed from different researchtreams and that the formats and structuring of the scales isnconsistent. The JIT-purchasing scale is Likert-based, requiringespondents to indicate degree of agreement. The JIT-productioncale is categorically structured requiring respondents to indicatehether or not their organizations have adopted a particular JIT-roduction practice. In future research efforts, we recommend thathe scales be reformulated for consistency.

The study focused on large U.S. manufacturers because we felthis group is more likely to have adopted JIT and agile practices. Asresult, it may not be appropriate to generalize results to mediumnd small manufacturers. Further, the theory as developed andested applies only in the manufacturing sector. Caution should bexercised when generalizing the results to the service and govern-ental sectors.

.2. Future research

This study links JIT practices to manufacturing agility andgility to performance. Additional research aimed at verifyinghese results is necessary. Sample frames that focus on small and

edium-sized manufacturers are necessary to facilitate general-zation of these results. It may also be advantageous to view theombination of JIT-purchasing (with JIT-production as antecedent)nd agile manufacturing as an overall supply chain strategy. Also,urther research could incorporate the other elements of lean

anufacturing such as TQM, preventive maintenance and humanesource management. Once the impact of each element haseen evaluated comparisons can be made between the effectsf individual elements compared to the effect of all elementsorking “synergistically” [Shah and Ward’s (2003) term applied

Management 29 (2011) 343–355

to the four bundles of lean manufacturing working in concert].Although uncertainty was not included in our models, concernthat it may moderate the hypothesized relationship between agilemanufacturing and operational performance led to subsequenttesting for moderation external to our models. Results indicatedthat environmental uncertainty did not moderate the relation-ship between agile manufacturing and organizational performancesupporting our original determination not to include it in theanalysis. Details of the assessment for moderation are presentedin Appendix B. While environmental uncertainty did not play asignificant role in our models (see Appendix B), future studiescould be expanded by including market environment (degree ofturbulence or degree of uncertainty) as a variable. It would beinformative to examine the degree of “match” between environ-ments (stable/lean vs. turbulent/agile) and determine how “degreeof match” impacts operational, financial and marketing perfor-mance.

Finally, the study could be strengthened by the inclusion ofitems that determine if the home plant of the respondent, usuallya plant manager, is one of multiple plants in an organization. Thisknowledge may lead to further interesting analysis of multi-plantfirms where some plants are focused on lean and others on agility.

Appendix A. Measurement scales

Agile manufacturing (alpha = .85)Note: Items 6, 8 and 10 removed to achieve unidimensionality.Please indicate the extent to which you agree or disagree with each

statement. (1 = strongly disagree, 7 = strongly agree)

1. This organization has the capabilities necessary to sense, per-ceive and anticipate market changes.

2. The production processes of this organization are flexible interms of product models and configurations.

3. This organization reacts immediately to incorporate changesinto its manufacturing processes and systems.

4. This organization has the appropriate technology and techno-logical capabilities to quickly respond to changes in customerdemand.

5. This organization’s strategic vision emphasizes the need forflexibility and agility to respond to market changes.

6. This organization has formed co-operative relationships withcustomers and suppliers.

7. This organization’s managers have the knowledge and skillsnecessary to manage change.

8. This organization has the capabilities to meet and exceed thelevels of product quality demanded by its customers.

9. This organization has the capabilities to deliver products to cus-tomers in a timely manner and to quickly respond to changesin deliver requirements.

10. This organization can quickly get new products to market.

JIT-Purchasing (alpha = .79)Note: Item 7 removed to achieve unidimensionality.Please indicate the extent to which agree or disagree with each

statement (1 = strongly disagree, 7 = strongly agree).

3. Production plans are shared with suppliers.4. Small lot size orders are placed with suppliers.5. Inspection of incoming materials has been reduced.6. Our staff visits suppliers’ plants on an informal basis.7. We involve suppliers in new product/materials design.


tions

m

11111111

r

aw

Following the general methodology described by Baron andKenny (1986), moderation is assessed. When testing for modera-tion, it is desirable that the moderator variable (EU) be uncorrelatedwith the predictor (AM) and criterion (OP) variables (Baron andKenny, 1986). Descriptive statics are displayed in Table B1. The

Table B1Descriptive statistics.

Mean Standard deviation

Agile manufacturing 4.89 .92Operational performance 5.26 .69Environmental uncertainty 3.61 1.06

Table B2Correlations.

TC

D

TC

D


JIT-Production (KR20 = .818, alpha based on composites = .78)Item 7 was removed based on KR20 assessment.Please indicate which of the following JIT practices have been imple-

ented in your organization’s production processes.

1. Kanban2. Integrated product design3. Integrated supplier network4. Plan to reduce setup time5. Quality circles6. Focused factory7. Preventive maintenance8. Line balancing9. Education about JIT0. Level schedules1. Stable cycle rates2. Market-paced final assembly3. Group technology4. Program to improve quality (Product)5. Program to improve quality (Process)6. Fast inventory transportation system7. Flexibility of worker’s skill

Operational performance (alpha = .80)Note: The cost management and asset management factors were

emoved to achieve unidimensionality.Please rate your company’s performance in each of the following

reas as compared to the performance of your competitors. (1 = muchorse than competition, 7 = much better than competition)

Customer service1. Customer satisfaction2. Product customization3. Delivery speedCost management4. Logistics costQuality5. Delivery dependability6. Responsiveness7. Order flexibility8. Delivery flexibility
Productivity
9. Information systems support10. Order fill capacity11. Advance ship notificationAsset management

able B3oefficients for agile manufacturing, environmental uncertainty, and interaction.

Model Unstandardized coefficients Standardized coe

B Std. error Beta

1 (Constant) 2.591 1.119AM .496 .216 .656EU .262 .311 .399AM × EU −.040 .060 −.372

ependent variable: op.

able B4oefficients for agile manufacturing and interaction.

Model Unstandardized coefficients Standardized coe

B Std. Error Beta

1 (Constant) 3.490 .342AM .327 .080 .432AM × EU .009 .011 .086

ependent variable: op.

Management 29 (2011) 343–355 353

12. Inventory turn13. Return on assets

Financial Performance (alpha = .92)Please rate your organization’s performance in each of the following

areas as compared to the industry average. (1 = well below industryaverage; 7 = well above industry average)

1. Average return on investment over the past 3 years.2. Average profit over the past 3 years.3. Profit growth over the past 3 years.4. Average return on sales over the past 3 years.

Marketing Performance (alpha = .91)Please rate your organization’s performance in each of the following

areas as compared to the industry average. (1 = well below industryaverage; 7 = well above industry average)

1. Average market share growth over the past 3 years.2. Average sales volume growth over the past 3 years.3. Average sales (in dollars) growth over the past 3 years.

Appendix B. Moderating impact of environmentaluncertainty

AM OP EU

Agile manufacturing (AM) 1 .477** .038Operational performance (OP) .477** 1 .105Environmental uncertainty (EU) .038 .105 1

** Significant at the 0.01 level (2-tailed).

fficients t-value Significant Collinearity statistics

Tolerance VIF

2.315 .0232.296 .024 .101 9.871

.844 .401 .037 27.096−.672 .503 .027 37.111

fficients t-value Significance Collinearity statistics

Tolerance VIF

10.202 .0004.065 .000 .729 1.372

.812 .419 .729 1.372


354 R.A. Inman et al. / Journal of Operations

Table B5Environmental uncertainty adapted from Miller and Dröge (1986).

Please indicate the extent to which you agree or disagree with each statement.(1 = strongly disagree, 7 = strongly agree)

1. This organization must change its marketing practices frequently2. The actions of this organization’s competitors are unpredictable3. The demands and tastes of this organization’s customers are almost

unpredictable

cspTi(pppsacaa

R

A

A

B

B

B

B

B

C

C

C

D

F

F

G

G

G

G

G

G

4. It is necessary to frequently make major changes in this organization’sproduction processes

5. This organization’s products become obsolete at a rapid rate

orrelations are presented in Table B2 below indicate that EU is notignificantly correlated with either AM or OP. Moderation is sup-orted if the interaction (XY) is significant (Baron and Kenny, 1986).he results of regressing AM, EU, and XY against OP are presentedn Table B3. While the regression coefficient for the interactionXY) is not significant, it should be noted that multicolinearity isresent making the coefficients difficult to interpret. Table B4 dis-lays the results of regressing AM and XY against OP without EUresent in the model. Multicolinearity is not present. The regres-ion coefficient for XY is .009 with an associated t-value of .812 andcomputed significance level of .419. Based on these results, it is

oncluded that EU does not moderate the relationship between AMnd OP. EU is measured using a 5-item scale adapted from Millernd Dröge (1986) displayed in Table B5.

eferences

hire, S.L., Golhar, D.Y., Waller, M.A., 1996. Development and validation of TQMimplementation constructs. Decision Sciences 27 (1), 23–56.

rmstrong, J.S., Overton, T.S., 1977. Estimating nonresponse bias in mail surveys.Journal of Marketing Research 14 (3), 396–402.

aron, R.M., Kenny, D.A., 1986. The moderator-mediator variable distinction insocial psychological research: conceptual, strategic, and statistical considera-tions. Journal of Personality and Social Psychology 51 (6), 1173–1182.

ooth, R., 1996. Agile manufacturing. Engineering Management Journal 6 (2),105–112.

owersox, D.J., Closs, D.J., Stank, T.P., Keller, S.B., 2000. How supply chain competencyleads to business success. Supply Chain Management Review 4 (4), 70–78.

rox, J.A., Fader, C., 2002. The set of just-in-time management strategies: anassessment of their impact on plant-level productivity and input-factor sub-stitutability using variable cost function estimates. International Journal ofProduction Research 40 (12), 2705–2720.

yrne, B.M., 1998. Structural Equation Modeling with LISREL, PRELIS, and SIMPLIS.Lawrence Erlbaum Associates, New Jersey.

hristopher, M., Towill, D., 2001. An integrated model for the design of agile supplychains. International Journal of Physical Distribution & Logistics Management31 (4), 235–246.

laycomb, C., Dröge, C., Germain, R., 1999a. The effect of just-in-time with customerson organizational design and performance. International Journal of LogisticsManagement 10 (1), 37–58.

laycomb, C., Germain, R., Dröge, C., 1999b. Total system JIT outcomes: inventory,organization and financial effects. International Journal of Physical Distributionand Logistics 29 (10), 612–630.

wyer, F.R., Welsh, M.A., 1985. Environmental relationships of the internal politicaleconomy of marketing channels. Journal of Marketing Research 22 (4), 397–414.

reeland, J.R., 1991. A survey of just-in-time purchasing practices in the UnitedStates. Production and Inventory Management Journal 32 (2), 43–49.

rohlich, M.T., Westbrook, R., 2001. Arcs of integration: an international study ofsupply chain strategies. Journal of Operations Management 19, 185–200.

arver, M.S., Mentzer, J.T., 1999. Logistics research methods: employing structuralequation modeling to test for construct validity. Journal of Business Logistics 20(1), 33–57.

erbing, D.W., Anderson, J.C., 1988. An updated paradigm for scale develop-ment incorporating unidimensionality and its assessment. Journal of MarketingResearch 25 (2), 186–192.

ermain, R., Dröge, C., 1997. Effect of just-in-time purchasing relationships on orga-nizational design, purchasing department configuration, and firm performance.Industrial Marketing Management 26 (2), 115–125.

oldman, S.L., Nagel, R.N., 1993. Management, technology and agility: the emer-gence of a new era in manufacturing. International Journal of Technology
Management 8 (1–2), 18–38.
oldsby, T.J., Griffin, S.E., Roath, A.S., 2006. Modeling lean, agile, and leagile supplychain strategies. Journal of Business Logistics 27 (1), 57–80.

reen Jr., K.W., Inman, R.A., 2005. Using a just-in-time selling strategy to strengthensupply chain linkages. International Journal of Production Research 43 (16),3437–3453.

Management 29 (2011) 343–355

Green Jr., K.W., McGaughey, R., Casey, K.M., 2006. Does supply chain managementstrategy mediate the association between market orientation and organiza-tional performance? Supply Chain Management: An International Journal 11(5), 407–414.

Green Jr., K.W., Medlin, B., Whitten, D., 2004. Developing optimism to improve per-formance: an approach for the manufacturing sector. Industrial Management &Data Systems 104 (2), 106–114.

Gunasekaran, A., 1999a. Design and implementation of agile manufacturing sys-tems: editorial. International Journal of Production Economics 62, 1–6.

Gunasekaran, A., 1999b. Agile manufacturing: a framework for research and devel-opment. International Journal of Production Economics 62, 87–105.

Gunasekaran, A., Lai, K.-H., Cheng, T.C.E., 2008. Responsive supply chain: a com-petitive strategy in a networked economy. Omega: The International Journal ofManagement Science 36, 549–564.

Hair, J.F., Anderson, R.E., Tatham, R.L., Black, W.C., 1998. Multivariate Data Analysis,5th ed. Prentice Hall, New Jersey.

Harrison, A., 1997. From leanness to agility. Manufacturing Engineer 76 (6), 257–260.Hormozi, A.M., 2001. Agile manufacturing: the next logical step. Benchmarking: An

International Journal 8 (2), 132–143.Jin-Hai, Li, Anderson, A.R., Harrison, R.T., 2003. The evolution of agile manufacturing.

Business Process Management 2 (9), 170–189.Katayama, H., Bennett, D., 1999. Agility, adaptability and leanness: a comparison of

concepts and a study of practice. International Journal of Production Economics60–61, 43–51.

Kidd, P.T., 1994. Agile Manufacturing: Forging New Frontiers. Addison–Wesley,Reading, MA and Workingham.

Kline, R.B., 1998. Principles and Practice of Structural Equation Modeling. GuilfordPress, New York.

Kohli, A.K., Jaworski, B.J., 1990. Market orientation: the construct, research proposi-tions, and managerial implications. Journal of Marketing 54 (2), 1–18.

Koufteros, X.A., 1999. Testing a model of pull production: a paradigm for manu-facturing research using structural equation modeling. Journal of OperationsManagement 17 (4), 467–488.

Krishnamurthy, R., Yauch, C.A., 2007. Leagile manufacturing: a proposed corporateinfrastructure. International Journal of Operations and Production Management27 (6), 588–604.

Lambert, D.M., Harrington, T.C., 1990. Measuring nonresponse bias in customer ser-vice mail surveys. Journal of Business Logistics 11 (2), 5–25.

Maskell, B., 2001. The age of agile manufacturing. Supply Chain Management 6 (1),5–11.

Mason-Jones, R., Naylor, B., Towill, D.R., 2000. Lean, agile or leagile? Matching yoursupply chain to the marketplace. International Journal of Production Research38 (17), 4061–4070.

McCullen, P., Towill, D., 2001. Achieving lean supply through agile manufacturing.Integrated Manufacturing Systems 12 (7), 524–533.

Mehra, S., Inman, R.A., 1992. Determining the critical elements of JIT implementa-tion. Decision Sciences 23 (1), 160–174.

Miller, D., Dröge, C., 1986. Psychological and traditional determinants of structure.Administrative Science Quarterly 31 (4), 539–560.

Nahm, A.Y., Vonderembse, M.A., Koufteros, X.A., 2003a. The impact of organizationalculture on time-based manufacturing and performance. Decision Sciences 35(4), 579–607.

Nahm, A.Y., Vonderembse, M.A., Koufteros, X.A., 2003b. The impact of organiza-tional structure on time-based manufacturing and plant performance. Journalof Operations Management 21 (3), 281–306.

Narasimhan, R., Swink, M., Kim, S.W., 2006. Disentangling leanness and agility: anempirical investigation. Journal of Operations Management 24, 440–457.

Naylor, J.B., Naim, M.M., Berry, D., 1999. Leagility: integrating the lean and agilemanufacturing paradigms in the total supply chain. International Journal ofProduction Economics 62, 107–118.

Patterson, K.A., Grimm, C.M., Corsi, T.M., 2004. Diffusion of supply chain technolo-gies. Transportation Journal 43 (3), 5–23.

Podsakoff, P.M., Organ, D.W., 1986. Self-reports in organizational research: problemsand prospects. Journal of Management 12 (4), 531–544.

Ramesh, G., Devadasan, S.R., 2007. Literature review on the agile manufactur-ing criteria. Journal of Manufacturing Technology Management 18 (2), 182–201.

Richards, C.W., 1996. Agile manufacturing: beyond lean. Production and InventoryManagement Journal 37 (2), 60–64.

Robertson, M., Jones, C., 1999. Application of lean production and agile manufac-turing concepts in a telecommunications environment. International Journal ofAgile Manufacturing Systems 1 (1), 14–16.

Sarkis, J., 2001. Benchmarking for agility. Benchmarking: An International Journal 8(2), 88–107.

Schumacker, R.E., Lomax, R.G., 1996. A Beginner’s Guide to Structural Equation Mod-eling. Lawrence Erlbaum Associates, Inc., New Jersey.

Shah, R., Ward, P.T., 2007. Defining and developing measures of lean production.Journal of Operations Management 25, 785–805.

Shah, R., Ward, P.T., 2003. Lean manufacturing: context, practice bundles, and per-formance. Journal of Operations Management 21, 129–149.

Sharifi, H., Zhang, Z., 2001. Agile manufacturing in practice: application of a method-ology. International Journal of Operations & Production Management 21 (5–6),772–794.

Tan, K.C., Lyman, S.B., Wisner, J.D., 2002. Supply chain management: a strategic per-spective. International Journal of Operations and Production Management 22(6), 614–631.


tions

V

W

Y

Zhang, Z., Sharifi, Hossein, 2007. Towards theory building in agile manufacturingstrategy – a taxonomical approach. IEEE Transactions on Engineering Manage-


azquez-Bustelo, D., Avella, L., Fernandez, E., 2007. Agility drivers, enablers andoutcomes. International Journal of Operations and Production Management 27
(12), 1303–1332.
isner, J.D., Leong, G.K., Tan, K.-C., 2005. Principles of Supply Chain Management:A Balanced Approach. Southwestern, United States.

usuf, Y.Y., Adeleye, E.O., 2002. A comparative study of lean and agile manufacturingwith a related survey of current practices in the UK. International Journal ofProduction Research 40 (17), 4545–4562.

Management 29 (2011) 343–355 355

ment 54 (2), 351–370.Zhang, Z., Sharifi, H., 2000. A methodology for achieving agility in manufacturing

organizations. International Journal of Operations and Production Management20 (4), 496–512.

group 6 - relation to jit, operational performance and firm performance

Documents