Investigating Optimum Length of Physical Queues at Businesses and Its Impact on Customers’ Decision to Join such Queues

Queue is regarded as the central aspect of service organisations. Effective queue management in businesses has a high impact on customer behaviour and service operations. Understanding the behaviour of people when they look at the length of the physical queue is vital for the managers of service organisations on how to keep the queue at the optimum length to attract customers. It also helps the operations managers to determine the number of servers and staff in each period by minimising costs and making more profit. This thesis empirically investigates the impact of physical queue on the customer and the optimum length of the physical queue in service organisations to send a positive signal for customers to join the queue. In doing so, this research aims to understand the effect of queue on service variables and the optimum length of the queue in restaurants.
The focus of past research studies was on minimising the queue through mathematical calculations; it fails to identify the advantages of the queue and find empirically determine the right length of queues in service organisations. This research contributes to the queueing literature in several ways; first, understanding the impact of physical queue length on the service attributes like quality of services to attract customers in the business. It also provides the service operations managers with specific guidelines on how to manage the length of physical queues in their businesses.
To fill these gaps in the literature, the optimum length of physical queues is examined and analysed in different service industries. For this purpose, firstly, the questionnaire-based survey is carried out to test the theoretical framework to understand the interaction of the physical queue length, type of customers and business locations on the quality, offers and intention to switch to alternatives as service variables. Self-completion questionnaire is distributed to collect data online and hardcopy from a random sample of 1515 service consumers in the UK. Factorial MANOVA is adopted to analyse data and examine the relationship between variables. The findings show that under different queue lengths, customers evaluate the service variables differently. Secondly, to find the optimum length of physical queues for restaurants in the UK, this thesis used the mathematical approach of queueing theory to calculates the efficiency parameters of the right model. Based on the queueing model (M/M/1) in the restaurants, data are collected empirically through observation of four restaurants in urban and suburban areas for seven days in each restaurant. Data analysed according to queuing theory formulas and principles to find the optimum length of the physical queue at different times and days for each restaurant. Our findings from quantitative analyses of survey and observations show the effect and optimum length of physical queue on customers in different areas.
A comparison of three service industries demonstrates the relationship and interaction of variables. It shows that in the entertainments industry, quality is statistically significant based on the type of customer (p = .047) Queue length has just interaction in the foods service industry with the quality and offer variables. In all three service industries, availability-of-alternatives are statistically significant, when a different type of customer, business locations and length of the queue have interaction together.
In addressing the research aim, the results of observations show that in the suburban areas there are not any difference-by by adding more server to the system in the length of queue on Weekdays, as in all models there are maximum one customer on average in the queue. In Weekends, it shows that on the busy times when the length of the queue is more than expected times, by running one server to the system the length of the queue would be in the limited optimum length. The analysis of queueing models in the urban locations shows the length of queue on Weekdays reduced from 2.151 in a single server to 0.119 in two serves. It demonstrates that in all conditions one server is enough to run customers in the queue as it does not change customers’ perception and also helps to keep the physical queue in the optimum length.