The Application of Tradtional Manufacturing Costs of Quality to the Service Environment

THE APPLICATION OF TRADTIONAL MANUFACTURING COSTS OF QUALITY TO THE SERVICE ENVIRONMENT: STRATEGIC IMPLICATIONS OF INCREASING QUALITY IN LABOR-INTENSIVE SERVICE ENVIRONMENTS

Kevin Watson

School of Management, Marist College, Poughkeepsie, NY 12601 (845) 575-3000

Tony Polito

School of Business, East Carolina University, Greenville, NC 27858 (252) 328-6569

ABSTRACT

Under traditional Total Quality Management (TQM) theory, product and process improvements result in high net present value investments. Underlying this theory is the assumption that investment in prevention practices reduce internal, external, and appraisal costs; allowing companies to improve quality while reducing the cost of production as the quality level approaches zero defective. Implicit to this argument is the assumption that increased quality reduces rework, thus reduces the hidden costs of unproductive capacity, and that this capacity can be used to increase throughput. This theory works well when used for manufacturing operations; however, it is unclear whether the same assumptions hold true for service operations. A simulation of a high labor-intensive service environment is produced to explore the implications of improved quality to the service environment.

INTRODUCTION

“Application of the principles of total quality management (TQM) is proving attractive to major service industries in this country, in the expectation that this will help them to deliver a better quality service and achieve greater customer satisfaction” [6]. However, an investigation of the relevant literature shows that the emphasis of quality research has primarily been the traditional manufacturing environment; there is little to guide implementation of TQM in the service environment. It is clear that application of Just-In-Time/Total Quality Management (JIT/TQM) improves company performance through reduction of costs and increased sales in the manufacturing environment:

“Continual reduction in mistakes, continual improvement in quality, mean lower and lower costs...less waste of materials, machine tie, human effort...As costs go down, through less rework, fewer mistakes, less waste, your productivity goes up...better quality and lower price, with a little ingenuity in marketing, will create a market.” [31]

These benefits have been statistically, and more importantly financially, proven [1]. However, service practitioners have yet to experience the same degree of cost savings, increased profitability, or increased throughput that manufacturers have through implementation of TQM practices. This begs the question: are manufacturing TQM strategies and practices applicable to the service environment? This paper will attempt to answer a portion of this question by investigating the underlying assumption of TQM, that increased quality reduces variance, reduces costs, and improves throughput in service operations. In order to do so, a simulation of a high labor-intensive service environment is produced to explore the implications of improved quality to the service environment.

According to Flynn et al. [12], the elements of a TQM system can be broken into common JIT and TQM infrastructure components and TQM unique practices. Through implementation of these practices, companies are able to decrease variability within their production system, thereby decreasing cycle time, reduce waste, and improve quality. These practices have a synergistic effect and must be integrated into a company in totality in order to assure improved performance. Of particular interest for service organizations are the management support and workforce management practices. For instance, management support is essential in order to assure the unifying vision necessary in order to assure that all employees are focused on customer needs, a necessary requirement in order to match customer expectations. [5] [27] Additionally, management support is necessary in order to assure that sufficient human resource capabilities are budgeted in order to sustain higher recruitment and training requirements, necessary in order to improve employee performance and retention under a TQM system. Typically, superior recruiting costs an additional $20-30 per employee [15] and training a single employee can exceed $35 per hour [20]. In addition, management must be willing to pay employees above average wages in order to retain them [14]. These costs are easily measured using traditional accounting practices; however, much like implementation of management information systems, their benefits are much harder to account for [9] [29].

Implicit in the argument for zero defects [8] and the rational for implementation of a TQM production philosophy, is the large percentage of resource utilized for the correction of defective items. This so called “hidden plant” [11] is a potential source of additional revenues in manufacturing, as capacity can be used in production of additional output for the same level of resource. Many authors have found similar wasted capacity in service organizations [3] [4] [10] [18] [22] or losses in future revenue [17] [21] [26] due to substandard quality. However, due to the nature of the service environment is such that services must be under-capacity scheduled in order to meet peak periods of customer demand, the manufacturing paradigm may not be applicable to the service environment. In other words, because of the perishable nature of services [13], the production and consumption of services is instantaneous, the reduction of unproductive capacity in the form of defective service delivery may only result in additional free time for the worker and not increased competitiveness for the company.

SIMULATION

Two simulations with differing quality levels were designed using the SLAM II language. The purpose of these simulations was to determine the level of resource utilization for both a baseline and improved service quality case. Implicit in the design of the simulations is the assumption that quality is a function of service speed, which can be improved through greater training of customer service representatives. The use of resource utilization as a measure of service quality is consistent with Heskett’s [19] research in the context of traditional queuing theory as set forth by Maister [24]. Sim 1 (baseline) and Sim 2 (improved service) were designed to establish a baseline for resource utilization for a complex high labor-intensive service environment with regeneration. The simulations contain two independent resources providing a single service given a normally distributed service time. The use of level resource capacity, rather than the use of flex time, is consistent with the literature on customer-driven interaction [2] [28]. Customers enter the simulation through a create node every 120 minutes based on a Poisson distribution. The entry time of 120 minutes was determined based on the need to create new customers throughout the run time of the simulation and the constraint of 300 separate entities placed upon the system by the computer language. Customers are originally routed through an assign node, which establishes their original expectations for the service encounter. This number was originally assigned a value of 4, representing medium level of expectation. After the customers pass through this assign node, they enter a single queue with a maximum size of six with zero customers originally in the queue. Should the queue reach its maximum number of customers, new entries are placed into the system at an assign node that decreased the level of satisfaction with the service experience and then returned to the system after a short delay. The service time of 3 minutes with a standard deviation of 20 seconds was established via a field study and represents the mean service time and standard deviation of an inside teller at a local bank and adjusted to allow for the constraints placed upon the system by the computer language. Upon completion of the service encounter, all customers enter a series of assign nodes established to improve, decrease, or leave the same the level of satisfaction with the service. There was a 16 percent chance for either improvement (+1) or disappointment (-1) and a 68 percent chance for the level of satisfaction to remain unchanged. The customer was than returned to the system at a point prior to the queue after a short delay if the value representing satisfaction was greater than 1. If the level of satisfaction fell to 1, the customer exited the system after a series of information was obtained. The simulation was run for the equivalent of 50 weeks of continuous service. Sim 2 is essentially the same simulation as Sim 1 with two exceptions. First, initial customer expectations for the service encounter are assigned a value of 3, representing heightened level of expectation. Second, the service time was reduced to 2 minutes with a standard deviation of 10 seconds was established via regression using data from a field study and represents a 33% improvement in service speed.

DISCUSSION OF SIM 1 & SIM 2

Sim 1 established the baseline for the service environment. Following the guidelines for utilization established by Heskett [19], the entry into the system was manipulated until the utilization rate of the resource was determined to be approximately 75 percent.

After ten independent runs to establish the validity of the system, Sim 2 was developed to determine the resource utilization given improved quality. Given the improved quality levels, measured by improvements in speed and decreased standard deviation, the level of resource utilization fell to 63 percent for the two resources in the system. Ten independent runs at the new service rate established the validity of these numbers. According to Heskett [19], this level of resource utilization would be sufficient to negatively influence the perceived overall quality level of the system. As such, improved training does not result in improved customer quality perception and may actually result in lower perceived quality.

Among the additional data obtained from the customers exiting the system were variables representing the number of times a customer had successfully passed through and the total time that the customer had spent in the system. The mean value for the number of customer turns significantly decreased from Sim 1 to Sim 2; however, the mean time in the system increased slightly. The mean number of turns decreased from 317 to 206, representing the problems experienced in trying to meet the heightened expectation level for the service of the customer. The mean value for the time in the system increased from 58,045 to 59,147, representing a slight increase. This increase is unexpected, especially in light of the decrease in the mean number of turns. However, following the interpretation of a histogram, this increase can be explained through increased satisfaction of the service if the customer’s original expectation level can be achieved.

CONCLUSION

Analysis of the simulations lends credence to the idea that the application of the manufacturing quality paradigm to the service environment is flawed. This is in agreement with Crosby’s assertion that “when service is discussed the focus is on people with the thought that a service is less easily described and measured which means that a different set of concepts for managing quality will be needed.”[7] While the application of additional training to the workforce does reduce service time, this additional available capacity does not result in additional throughput and is wasted in the high labor-intensive service environment due to the customer dependent nature of production/delivery.

Combining Crosby’s "absolutes of quality" [8] with Meister’s [25] definition of service quality, we can see how the quality paradigm must be changed to reflect the service environment: Quality is exceeding customer expectations and the measurement of quality is the NPV of all future revenue streams due to non-conformance. Therefore, the strategic implication of TQM in the service environment may lie in its potential to differentiate a service provider from its competitors [16] [23] and thereby gain customer loyalty and improve profits [13] [26] [30].

REFERENCES

[1]Anonymous. 1991. Management practices: U.S. companies improve performance through quality efforts. United States General Accounting Office. U.S. Government Printing.

[2]Armistead, C. and G. Clark, 1994. The coping capacity management strategy in services and the influence on quality performance. International journal of service industry management. 5(2).

[3]Avkiran, N. Developing an instrument to measure customer service quality in branch banking. 1994. International journal of bank marketing. 12(6).

[4]Barnes, J. and J. Cumby. 1993. The cost of quality in service-oriented companies: making better customer service decisions through improved cost information. Proceedings of the 1993 ASB Conference.

[5]Belton, E. 1993. Cost cutting: Battleground of the 1990s, Part Two. Canadian underwriter. 60(9): 2426.

[6]Cowling, A. and K. Newman. 1995. Banking on people: TQM, service quality, and human resources. Personnel review. 24(7).

[7]Crosby, P. 1996. Thinking about excellence: Service is a product. Journal for quality and participation. 19(6).

[8]Crosby, P. 1979. Quality is free. McGraw-Hill. New York, New York.

[9]Fanjoy, B. 1994. Bringing financial discipline to service quality. TQM Magazine. 6(6).

[10]Feigenbaum, A. 1993. Feigenbaum's window on the world: Regaining the quality service edge. National Productivity Review. 12(4): 457-461.

[11]Feigenbaum, A. 1983. Total quality control. Third edition. McGraw-Hill. New York, New York.

[12]Flynn, B., Sakakibara, S. and R. Schroeder. 1995. Relationship between JIT and TQM: Practices and performance. Academy of management journal. 38(5).

[13]Ghobadian, A., Speller, S. and M. Jones. 1994. Service quality: Concepts and models. International Journal of quality and reliability management. 11(9).

[14]Golhar, D. and S. Deshpande. 1993. An empirical investigation of HRM practices in JIT firms. Production and inventory management journal. 34(4): 28-32.

[15]Greengard, S. 1995. Are you well armed to screen applicants? Personnel journal. 74(12).

[16]Gronroos, C. 1978. A service oriented approach to marketing of service. European journal of marketing. 12(8).

[17]Grubbs, R. and E. Reidenbach. 1991. Customer service renaissance: Lessons from the banking wars. Probus. Chicago, Illinois.

[18]Harvey, T. 1995. Quality: The only profit strategy. Bank marketing. 27(1).

[19]Heskett, J. 1986. Managing in the service economy. Harvard Business School Press. Cambridge, Massachusetts.

[20]Hubbard, A. 1995. A price tag on training. Mortgage banking. 55(5).

[21]Kennedy, S. 1997. Waking up to the realities of customer satisfaction. CMA Magazine. 71(1).

[22]LeBlanc, G. and N. Nguyen. 1988. Customers' perceptions of service quality in financial institutions. International journal of bank marketing. 6(4).

[23]Lehtinen, U. and J. Lehtinen. 1992. Service quality: A study of quality dimensions. Working paper. Service Management Institute of Helsinki.

[24]Maister, D. 1985. The psychology of waiting lines. In Czepiel, J., Soloman, M. and C. Surprenant (editors). The service encounter. D.C. Heath & Company. Lexington, Massachusetts.

[25]Meister, J. 1990. Service marketing: Rewards reap results. Marketing news. 24(13).

[26]Raddon, G. 1987. Quality service: A low-cost profit strategy. Bank marketing. 19(9).

[27]Reynierse, J. 1993. Building the quality service focused bank. Bank marketing. 25(4).

[28]Sasser, W. 1976. Match supply and demand in service industries. Harvard business review. November-December.