Development and validation of resource flexibility measures for manufacturing industry

Purpose: Global competition and ever changing customers demand have made manufacturing organizations to rapidly adjust to complexities, uncertainties, and changes. Therefore, flexibility in manufacturing resources is necessary to respond cost effectively and rapidly to changing production needs and requirements. Ability of manufacturing resources to dynamically reallocate from one stage of a production process to another in response to shifting bottlenecks is recognized as resource flexibility. This paper aims to develop and validate resource flexibility measures for manufacturing industry that could be used by managers/ practitioners in assessing and improving the status of resource flexibility for the optimum utilization of resources. Design/methodology/approach: The study involves survey carried out in Indian manufacturing industry using a questionnaire to assess the status of various aspects of resource flexibility and their relationships. A questionnaire was specially designed covering various parameters of resource flexibility. Its reliability was checked by finding the value of Cronback alpha (0.8417). Relative weightage of various measures was found out by using Analytical Hierarchy Process (AHP). Pearson’s coefficient of correlation analysis was carried out to find out relationships between various parameters. Findings: From detailed review of literature on resource flexibility, 17 measures of resource flexibility and 47 variables were identified. The questionnaire included questions on all these


Introduction
The approach to manufacturing has undergone a considerable change in the past two decades or so. In today's business world, emphasis is shifting from mass production of low-cost, interchangeable commodities to the production of high-quality goods and services made individually or in small batches to meet the specific demands of small groups of consumers.
This shift requires greater flexibility in manufacturing system to accommodate rapid changes in product design as per consumer demand (Wagner & Hollenbeck, 2010). Today's market is determined by customers. For producers to exist, they must seek and produce what potential consumers require. The customers, these days, have many options available and only that product, which comes up to a customer's expectations may dominate the market. Industrial systems have become very complex owing to a large variety of products being made in a single manufacturing firm. A number of different types of materials, machines, tools, skill levels, and other inputs have to be employed in a production system. Market uncertainties, because of a scarcity of resources and rapid product innovations, add to the decision-making complexities in the manufacturing system (Zapfel, 1998). Achieving higher levels of productivity in this complex environment requires a system to rapidly adjust itself to complexities, uncertainties, and changes. Thus, flexibility is required for productivity.
Simply stated, flexibility is the ability of a system to react to and accommodate changes (Chauhan & Singh, 2011). To remain competitive, flexibility must exist during the entire life cycle of a product, from design to distribution (Chauhan, Singh & Sharma, 2010).
Manufacturing flexibility has become an exclusive expression that indicates a manufacturing system's ability to respond to fluctuations in the production process and produce customeroriented products at low cost and greater response sensitivity in dynamically changing manufacturing systems. Flexibility is an inherent attribute and an intangible asset of a manufacturing enterprise. Flexibility is difficult to understand and quantify and is expensive to build (Chauhan, Singh & Sharma, 2007). Knowledge about its own inherent flexibility helps an enterprise to manage it in a more effective manner towards the organizational performance improvement. Flexibility can be used not only for effectively managing the changes but also for enhancing performance of manufacturing systems (Chen & Adam, 1991).
To remain competitive, many companies have improved their production processes by introducing manufacturing flexibility (Cox, 1989). The manufacturing flexibility that can only be achieved with flexible resources is termed as resource flexibility. Machines and workforce are the most important resources of a manufacturing organization (Chauhan & Singh, 2011). Thus, machine and labor flexibility forms the foundation blocks of manufacturing flexibility. Moreover, other types of flexibilities, such as process flexibility, operation flexibility, product flexibility, routing flexibility, and product-mix flexibility depend on labor and machine flexibility (Karuppan & Ganster, 2004). Resource flexibility has received an increasing amount of popularity in the past two decades as it provides companies with the ability to adhere to disturbances in the production process so that new and existing products can be produced more rapidly.
Resource flexibility follows the organization's community partnerships and quality improvement processes. Using flexibility in the allocation of financial, human and physical resources enables local decisions about objectives and strategies from the partnerships plan to be put into effect. Daniels, Mazzola and Shi (2004) have demonstrated that allocation of partial resource flexibility shows substantial improvements in operational performance in both serial and parallel-machine production environments through the effective utilization of resources. Process divergence and diverse process flows must be considered while taking decisions about resource flexibility. In case of immense task divergence and flexible process flow high level of resource flexibility is required. Accordingly, employees need to perform a broad range of duties, and equipment must be general purpose. Otherwise, utilization will be too low for economical operations (Krajewski, Ritzman & Malhorta, 2010). Thus, it seems to be important to gain an understanding of how resource flexibility can be managed in manufacturing systems. This paper is focused on the development and validation of resource flexibility measures in Indian manufacturing industry.
The paper is structured as follows: in section 2 the manufacturing flexibility definitions and technological attributes defined by previous literature are reviewed. In section 3, the methodology adopted for the study, resource flexibility construct and details of survey instrument have been presented. Section 4 presents the detailed analysis, results and impact of various measures on resource flexibility. Finally, in section 5, conclusions and guidelines for future research with regard to resource flexibility are developed.

Literature Review
The need and attempts to compute the value of flexibility may be viewed as a hedge against future uncertainty (Sethi & Sethi, 1990). Also, researchers need flexibility measures to test theories and operation managers need them to facilitate making capital investment decisions and in determining performance levels. Many of the existing flexibility studies have only investigated the concept of flexibility in relation to a particular domain and a specific objective, instead of considering an entire resource flexibility of the system. Gerwin (1987) has highlighted labor, machine, process and routing as the components of manufacturing flexibility. Sarker, Krishnamurthy and Kuthethur (1994) presented measures of machine, process and routing component level flexibility and concluded that it is quite impossible to have a universal scale of overall manufacturing flexibility. As a result, current flexibility models are simply based on a limited analysis of manufacturing systems (Koste & Malhotra, 1999). Therefore, while there are several taxonomies that attempt to define manufacturing flexibility, they are incomplete or too abstract to explain the fundamental concept of flexibility (Gupta & Buzacott, 1989;Shewchuk & Moodie, 1998). Thus, the meaning and implementation of overall resource flexibility still remains ambiguous (Chang, Whitehouse, Chang & Hsieh, 2001). An analytical model capable of generating a clear relationship between the degree of a system's flexibility and the level of a system's performance has yet to be defined (Slack, 1987;Kumar, 1987;Gupta & Goyal, 1989). Wahab (2005) proposed a generic model to measure machine and product mix flexibilities with consideration of uncertainties in the system. Manufacturing flexibility is the ability of the organization to manage production resources and uncertainty to meet various customer requests (Zhang, Vonderembse & Lim, 2003). Manufacturing flexibility is considered to be a strategic element of business, along with price (cost), quality, and dependability (Chauhan & Singh, 2011). According to Chen and Adam (1991), investment in flexible manufacturing systems leads to various advantages: less scrap, reduced downtime, improved quality, increased labor productivity, better machine efficiency and augmented customer satisfaction etc.
Machine flexibility is dependent on the ease with which one can make changes in order to produce a given set of part types (Browne, Dubois, Rathmill, Sethi & Stecke, 1984). A multi-skilled workforce is believed to enhance system performance (Treleven, 1989). Jaikumar (1989) has discovered that labor flexibility has proved to be the key to the success of flexible manufacturing systems in Japan. Machine flexibility is measured by the number of operations that a workstation performs and the time needed to switch from one operation to another (Tsourveloudis & Phillis, 1998). The extent of flexibility can be measured by its metrics; efficiency, responsiveness, versatility and robustness (Golden & Powell, 2000). Efficiency and versatility should be considered for the measurement of machine flexibility (Chang et al., 2001). Operators with a high level of skills should be assigned first and versatile operators last to maximize quality and minimize staffing costs (Franchini, Caillaud, Nguyen & Lacoste, 2001).
A flexible workforce is especially valuable in responding to the design changes and new product introduction. Higher labor flexibility provides enhanced capability to reassign tasks in the case of workforce absence (Singh, 2008).
Resource flexibility in the form of labor and machine flexibility can be judiciously exploited towards reduction in wastages in resources of manufacturing enterprise (Malhotra & Ritzman, 1990). Resource flexibility helps the firm to reduce manufacturing flow times, work-in-process inventories, and improve customer service while providing an efficient use of both labor and equipment (Polakoff, 1991). Substantial improvements in operational performance can be achieved through scheduling parallel manufacturing cells with effective utilization of resource flexibility (Daniels, Hoopes & Mazzola, 1996). Every operation may need several resources and furthermore, a resource may be selected from a given set of resources. Peres, Roux and Lasserre (1998) have presented multi-resource shop scheduling with resource flexibility by assigning operations to resources and sequence operations, in order to minimize the completion time. Depending on the amount of internal resources, a group may exit a market in response to increased competition, or channel funds to the subsidiary operating in that market.
Resource flexibility within a group makes subsidiaries' reaction functions flatter, thus discouraging rivals' strategic commitments when entry is accommodated (Cestone & Fumagalli, 2005

Research Methodology
The research methodology includes an extensive review of literature to identify various constructs of resource flexibility, determining the weightages of these constructs towards resource flexibility through the use of Analytic Hierarchy Process, survey of industry using a specially designed questionnaire which carries questions on various constructs and variables of resource flexibility and analysis of the response of the survey to calculate resource flexibility to validate the approach. The methodology adopted includes the following: • Development of resource flexibility construct from the review of literature • Collection of data on various measures through survey of manufacturing industry.
• Measurement of resource flexibility and validation of measures by statistical analysis.

Development of Resource Flexibility Construct
Many different types of flexibilities have been identified in the literature and the research indicates that the domain of any flexibility dimension is comprised of four elements: rangenumber, range-heterogeneity, mobility, and uniformity of a system or resource (Slack, 1983(Slack, , 1987Upton, 1994;Koste & Malhotra, 1999). In this study, the focus is on the overall resource flexibility measures that includes machines, labor and products (materials). These resources are frequently studied in the literature (Gupta & Somers, 1992;Nandkeolyar & Christy, 1992;Malhotra & Ritzman, 1990;Chauhan & Singh, 2011)

Assess the Weightage of Different Measures of Resource Flexibility
Although various measures, as listed above, contribute towards resource flexibility yet their contribution cannot be assumed equal. Weightage of some measures may be more than others. To determine their relative weightage, the analytical hierarchy process (AHP) was employed (Saaty, 1986;1990). Each measure is compared with other measures pair-wise. Three experts; one industrial manager, one senior production executive and one academician were involved in the process of paired comparison for determining the weights of various measures. They, however, filled the response in qualitative scale of very low, low, medium, high and very high as the difference between the importances of two measures. These qualitative responses are converted to the quantitative values using the scale as: very low = 1; low = 3; medium = 5; high = 7 and very high = 9. Position matrices were made, separately for each expert showing the paired comparison of each measure with the other measures. The weightage of each measure was determined by calculating an eigenvector and normalizing it for each expert's response. From the weightages of each measure, calculated in the above manner by each expert, mean weightage was calculated as shown in table 1. A consistency index (CI) and consistency ratio (CR) is also calculated to check the numerical and transitive consistency and validity of experts' judgments for resource flexibility measures. The most important measures of resource flexibility were found to be ''ability of machines to perform diverse set of operations'' and "ability of workers to work on different machines", with a contributing weightage of 20.19% and 17.58% respectively. These are followed by ''obsolescence rate of machines on introduction of new products'', ''skill level of workers to perform different jobs'', "reliability of machines over job change", "reliability of workers over job change", "ability of production workers to do autonomous maintenance" and "productivity effectiveness due to change of machine" with contributing weight of 9.99%, 8.57%, 7.09%, 5.35%, 5.35% and 5.17% respectively. Other measures have a contribution of less than five percent as shown in table 1. They are comparatively less important in the measurement of resource flexibility.

Design of Questionnaire
Questions were framed on all 17 constructs and 47 variables. Each question had a multiple choice answer and a seven point Likert scale ranging from strongly agree to strongly disagree with a middle point anchor of neither agree nor disagree. To ensure the relevance and effectiveness of questions to the industry, the questionnaire was pre-tested on a random sample of 12 units and the suggestions received are incorporated. Internal reliability of questionnaire items is tested by calculating Cronbach's alpha using the IBM SPSS 11.01 software. Cronbach's alpha is a coefficient of reliability. It is commonly used as a measure of the internal consistency or reliability and validation of measurement instruments such as questionnaires. It was first named alpha by Cronbach (1951). The survey questionnaire is found to be acceptable, with a Cronbach's alpha equal to 0.8417 (Radhakrishna, 2007).

The Survey of Manufacturing Industry
The survey was carried out in Indian manufacturing industry. A manufacturing firm is likely to reflect, to some degree all seventeen measures of resource flexibility included in the study.

Measurement of Resource Flexibility
Each question has seven options for the answer and thus a score between 1 and 7 is possible.
Thus each question can have a highest score of 7 and each construct a score of 7*n (where n is the number of variables in each measure). Actual score received by a construct is divided by the maximum possible score to calculate value of each measure on a scale of 0 to 1. To calculate the value of each measure from the raw scores of the questionnaire and the status of resource flexibility following equations were used: Where ∑ S a i is the sum of actual score of i th measure, which is further equal to:

Results and Discussion
The Pearson's coefficients of correlations between various measures of resource flexibility are worked out using SPSS 11.01. Pearson's coefficient of correlation (r) is a measure of the strength of the association between two variables. It can have a value anywhere between plus and minus one.
The larger the value of 'r', ignoring the sign, the stronger would be the association between the two variables and the more accurately one can predict one variable from the knowledge of the other. At its extreme, a correlation of 1 or -1 means that the two variables are perfectly correlated, -30-Journal of Industrial Engineering and Management -http://dx.doi.org/10.3926/jiem.655 meaning that one can predict the values of one variable from the values of the other variable with perfect accuracy. At the other extreme, zero value of 'r' implies an absence of correlation, i.e. there is no relationship between the two variables. The value of Pearson's coefficient of correlation between various measures and overall resource flexibility are presented in table 2. Table 2 shows that all correlations are positive, i.e. change in any one measure affects all the other measures and overall resource flexibility directly. A total of 272 correlations are determined, of which 224 emerged to be significant. Further, 192 of these correlations are significant at a level of p ≤ 0.01 and 32 at a level of p ≤ 0.05. This reflects that all measures of resource flexibility are complementary to each other. If an improvement is made in one the others also get improved. To validate the resource flexibility measures, their relationship with overall resource flexibility are also worked out and their relative impact on resource flexibility is shown in Figure 1. As can be seen, the main measures that influence the realization of resource flexibility are ''productivity effectiveness due to change of workforce'', "ability of workers to work on different machines", "ability of machines to perform diverse set of operations", "productivity effectiveness due to change of machine" and "cost effectiveness of workers over job change" with an impact factor of 78.2%, 76.3%, 72.8%, 70.5% and 70% respectively. It is surprising to find that "training of workers" is having a minimum impact of 9.3% to implement resource flexibility.

Conclusion
Resource flexibility is a polymorphous phenomenon visualized as a means to meet customers' demand quickly, provide a broad product range or introduce new products to the range easily.
The present study shows that most of the respondent firms have some level of resource flexibility. 'Productivity effectiveness due to change of manpower' and 'ability of workers to work on different machines' are found to be the leading measures of resource flexibility. Other measures which have been found to be at good level are ability of machines to perform diverse set of operations, productivity effectiveness due to change of machine, cost effectiveness of workers over job change, ease of machine setup or changeover, reliability of workers over job change, reliability of machines over job change, co-operation of workers in achieving production targets, attitude of workers towards change and obsolescence rate of machines on introduction of new products. However, the least practiced measure is training of workers, which indicates that companies do not want to spend on worker training.
In this study an approach has been developed involving 17 measures along with their weightages to build resource flexibility in a manufacturing organization. The approach has been applied through a survey of the industry and validated. The practitioners and managers can make use of the results of this study for managing resource flexibility in their organizations to survive in the present competitive scenario. Further, it was seen that all the measures are significantly correlated with overall resource flexibility except training of workers, as shown by Pearson's coefficient of correlation. The values of correlations provide guidelines to the manufacturing organizations to decide the hierarchy of measures for implementation. The study also concludes the human resource should be taken care off first, followed by machines and products for managing the resource flexibility in Indian manufacturing industry. The study has been limited to manufacturing industry. Future research can focus on resource flexibility in other areas like offices, finance, marketing, process industry and service industry..