Analyze Optimal queuing structure of crowd to utilize the given space.

Abstract

Managing crowd is one of the most critical and challengeable tasks in these days. When considering about the queues at ticketing counters, ATM machines and most of the other places people needs to wait after one or two at least in a waiting line. Waiting at a queue more than expected time is shows the inefficiency of the queue management. Controlling the queue according to the requirements may increase the equity by speedup the checkout process. This research has been done to introduce a solution for that. People can come up with a solution based on the best practices in human management as well as using computer technology. There are two main methods available to solve this problem in computer science domain to predict the best possible queue structure, Agent based and Grid based method. This research is based on the grid based method. The area, which is going to be utilized by the queue lines are converted into a grid/ mesh. Mesh is based on the average human area and each node in a mesh used to create the optimal structure of queue lines. Initially the entry points, service counter locations and the coordinates will be the inputs for the developed prototype. The area can be divided into several queue lines when there is a multiple service counters in the same area. Partitioning of the area is depend on the weight allocated for each service counters by considering the priorities and other properties. For example some service counters displayed as “Express checkout – only five or less number of unit per checkout”. That is prioritization and it is also a input to the system. System is predicting the optimal queuing structure based on the algorithms of path planning and area partitioning. Calculating the next possible node according to the rules defined it identified the traverse path from start to the end node. Before path planning it is necessary to calculate the correct region that can be taken for each and every queue lines according to the weights assigned. Neighbours are known and define the next possible neighbour of the same queue using the properties of current node is the concept behind the area partitioning. System will output the optimal path based on the grid units. Person who is interesting in creating a design for ticketing counter or any other crowd manages in terms of queues can used this tool effectively. Feedback to the questionnaire and test results shows the importance and the accuracy of the system.