E. Ruelas–Gonzalez*1, J. Limon–Robles2, N. Smith–Cornejo3
1 ITESM Campus Monterrey, Center for Quality and Manufacturing. CIAP 6th floor. 2501 Eugenio Garza Sada South Av., Monterrey, Nuevo León. Mexico. 64849. Phone: 8358.2000, 83581400. Ext. 5197. *E–mail: A00744107@itesm.mx.
2 ITESM Campus Monterrey. Engineering School. CIAP 6th floor. 2501 Eugenio Garza Sada South Av. Monterrey, Nuevo León. Mexico. 64849.
2 ITESM, Campus Monterrey. Engineering School. CEDES 4th floor. 2501 Eugenio Garza Sada South Av. Monterrey, Nuevo León, Mexico. CP. 64849.
ABSTRACT
A major concern for convenience store managers is lost sales due to balking. Convenience stores customers pay high margins expecting fast service. If waiting lines are too long for their tolerance level at their arrival, they balk and the sale is lost as a result. In order to reduce lost sales, the length of the waiting line is usually controlled by opening additional checkout registers when the number of customers standing in the line exceeds a specified number and maintaining them open until they are no longer needed . This paper presents an applied approach to model the probability that customers actually enter the store and define the optimal opening level (N–policy) of the second checkout register based on several factors including the particular waiting line length tolerance level of usual customers, the average hourly arrival rate of customers to the store and the average gain per customer transaction. Several performance measures are computed. The total expected cost function per unit time is proposed to determine the optimal operating N–policy at minimum cost. The model is applied in a real case of a convenience store chain.
]]> Keywords: Convenience store, balking process, customer service model, queuing control, removable servers, N–policy.
RESUMEN
Uno de los mayores retos que enfrentan los administradores de las tiendas de conveniencia o servicio rápido es la pérdida de clientes debida al fenómeno conocido como "balking". Los clientes pagan altos costos en éste tipo de tiendas esperando un servicio rápido así que su tolerancia a la espera no es muy alta. El "balking" se presenta cuando un cliente decide no llevar a cabo la compra del producto o servicio porque considera que la fila de espera, a su llegada, es demasiado larga para su tiempo disponible y paciencia. Con la finalidad de reducir las ventas perdidas, la fila de espera es controlada abriendo cajas adicionales cuando ésta excede un número determinado de clientes. La caja se mantiene abierta hasta que ya no hay más clientes esperando. Este artículo presenta una metodología aplicada para modelar la probabilidad de que el cliente decida ingresar a la línea de espera para pagar y así mismo, define la política de apertura óptima de la segunda caja (N–policy) basada en diversos factores incluyendo la tolerancia específica de los clientes típicos de éste tipo de servicios, la tasa promedio de llegadas por hora a la tienda y la ganancia promedio por venta al cliente. Se calculan varias medidas de desempeño y adicionalmente una función del costo total esperado es propuesta para determinar la política óptima de operación al mínimo costo. El modelo se aplica a una cadena de tiendas de conveniencia mexicana con gran presencia en el norte del país.
DESCARGAR ARTÍCULO EN FORMATO PDF
References
[1] J. Bae and S. Kim. Average cost under the p–lambda,tau(m) policy in a finite dam with compound poisson inputs. Journal of Applied Probability, 40:519.526, 2003. [ Links ]
]]>[2] C.E. Bell. Characterization and computation of optimal policies for operating an m/g/1 queueing systems with removable server. Operations Research, 19:208.218, 1971. [ Links ]
[3] K. Boots and H. Tijms. A multiserver queueing system with impatient customers. Management Science, 49:444.448, 1999. [ Links ]
[4] F. Budnick and R. Mojena. Principles of operations research for management. Richard D. Irwin, Inc., Homewood, IL., 1977. [ Links ]
[5] Erhan Çinlar. Introduction to Stochastic Processes. Prentice–Hall, Inc., Englewood Clins, N.J., 1975. [ Links ]
[6] M. Davis. How long should a customer wait for service? Decision Sciences, 22:421, 1991. [ Links ]
]]>[7] E. Feinberg and O. Kella. Optimality of d–policies for an m/g/1 queue with a removable server. Queueing Systems, 42:355.376, 2002. [ Links ]
[8] J. S. Fitzsimmons. Service Operations Management. McGraw Hill Co., New York, 1982. [ Links ]
[9] R. Gebhard. A queueing process with bilevel hysteretic service–rate control. Nav. Res. Log. Quart, 14:55.68, 1967. [ Links ]
[10] D. Gross and Carl M. Harris. Fundamentals of Queueing Theory. Wiley Series in Probability and Mathematical Statistics. John Wiley & Sons, Inc., second edition, 1985. [ Links ]
[11] R. Hassin. Consumer information in markets with random product quality: The case of queues and balking. Econometrica, 54:1185.1195, 1986. [ Links ]
]]>[12] D.P. Heyman. Optimal operating policies for m/g/1 queueing systems. Operation Research, 16:362.382, 1968. [ Links ]
[13] L.K. Jones. Inferring balking behavior from transactional data. Operations Research, 47:778, 1999. [ Links ]
[14] P. Naor. The regulation of queue size by levying tolls. Econometrica, 37:15.24, 1969. [ Links ]
[15] A.R. Pazgal. Comparison of customer balking and reneging behavior to queueing theory predictions: An experimental study. Computers and Operations Research, 35:2537.2548, 2008. [ Links ]
[16] M. Sobel. Optimal average–cost policy for a queue with start–up and shut–down costs. Operations Research, 17:145.162, 1969. [ Links ]
]]>[17] M. Van Tits. Simulation of a queueing problem with balking. ACM SIGSIM Simulation Digest, 11:58.64, 1979. [ Links ]
[18] K. Wang and Y. Chang. Cost analysis of a finite m/m/r queueing system with balking, reneging and server breakdowns. Mathematical Methods of Operations Research, 56:169.180, 2002. [ Links ]
[19] K. Y. Wang. Optimal control of an m/hk/1 queueing system with a removable server. Mathematical Methods of Operations Research, 57:255.262, 2003. [ Links ]
[20] M.N. Yadin. Queueing systems with a removable service station. Opl Res Q, 14:393.405, 1963. [ Links ]
Acknowledgments
]]> This research and divulgation has been supported by ITESM Campus Monterrey Research Fund CAT128. ]]>