A computer program product comprising a non-transitory computer readable medium having a computer readable program embodied therein, wherein the computer readable program, when executed on a computing device, causes the computing device to:
- receive an image of each queue of the plurality of queues, using at least one image acquisition devices, wherein each queue comprises a plurality of customers waiting in at least one queue of a plurality of queues with associated service items cart for a service;
- extract a set of sub-images for each customer among the plurality of customers, from the received image of each queue, using a pre-trained deep learning based recognition model, wherein the set of sub-images for each customer comprises a face of the customer and the associated service items cart, wherein the pre-trained deep learning based recognition model recognizes the associated service items cart and maps the associated service items cart with the customer, wherein the pre-trained deep learning based recognition model includes a pre-built demographics prediction model to estimate age of the each customer and an accuracy age estimate factor, based on the face of the customer, and wherein the accuracy age estimate factor is a percentage defining the accuracy of the prediction of the pre-built demographics prediction model;
- estimate visual cues of the plurality of customers such as the age and a senti-motional state of each customer from the face of the customer comprised in the corresponding set of sub-images, and a number of service items from the associated service items cart present in the corresponding set of sub-images, using the pre-trained deep learning based recognition model, wherein the number of service items of the customer is estimated based on whether the associated service items cart is at least one of full, half-full and empty, and wherein the visual cues capture facts of the customer from the plurality of customers waiting in the plurality of queues in real-time;
- determine a demographic factor (D) of each customer based on the corresponding estimated age, a senti-motional factor (S) of each customer based on the corresponding estimated senti-motional state, and a service items cart load factor (L) of each customer based on the estimated number of service items, wherein the demographic factor (D) of each customer is determined from a predefined set of demographic factors and each demographic factor of the predefined set of demographic factors is defined based on a predefined range of the estimated age, wherein the senti-motional factor (S) of each customer is determined from a predefined set of senti-motional factors and each senti-motional factor of the predefined set of senti-motional factors is defined based on a set of senti-motional states, wherein the set of senti-motional states includes a positive attitude and a negative attitude, wherein emotions of each customer include at least one of a happy state, a sad state and an irritation state, and wherein the service items cart load factor (L) of each customer is determined from a predefined set of service items cart load factors and each service items cart load factor of the predefined set of service items cart load factors is defined based on a predefined range of the estimated number of service items, and wherein the range of predefined set of demographic factors, the range of predefined set of senti-motional factors and the range of predefined set of service items cart load factors are defined based on historical transactions observed for a plurality of service areas;
- calculate a service time (T) for each customer, based on the corresponding demographic factor (D), the corresponding senti-motional factor (S) and an average efficiency (E) of a service operator, wherein the average efficiency (E) of the service operator is determined, based on an average time (E1) taken by the service operator to process a service item from a plurality of service items processed for the plurality of customers, an average time (E2) taken by the service operator to greet and pay of each customer, and the service items cart load factor (L) of each customer;
- and
- predict the wait time of each queue of the plurality of queues in real time, based on the service time of the plurality of customers waiting in the associated queue for the service wherein the wait time is predicted accurately by performing analysis of the prediction of wait time based on:
- a queue in which the customer with a sad face and an irritation face takes more time than the customer with a happy face during a check-out process;
- checking for the plurality of queues which are less crowded;
- checking for the customers shopping a smaller number of shopping items than the customers shopping a greater number of shopping items and pay using a variety of modes of payment;
- and
- checking for the customers taking more time during the check-out process than the customers taking less time.