Scheduling algorithm to select $k$ optimal programme slots in television channels: A graph theoretic approach

TL;DR

This paper models television programme scheduling as a maximum weight k-coloring problem on interval graphs, aiming to select mutually exclusive slots across channels to maximize viewers for advertising.

Contribution

It introduces a graph-theoretic model for scheduling and provides an efficient algorithm to optimize viewer reach for advertisements across multiple channels.

Findings

The problem is formulated as a maximum weight k-coloring on interval graphs.

An algorithm with complexity O(kMn^2) is proposed for optimal scheduling.

The approach effectively maximizes viewers for advertising in television scheduling.

Abstract

In this paper, it is shown that all programmes of all television channels can be modelled as an interval graph. The programme slots are taken as the vertices of the graph and if the time duration of two {programme slots} have non-empty intersection, the corresponding vertices are considered to be connected by an edge. The number of viewers of a programme is taken as the weight of the vertex. A set of programmes that are mutually exclusive in respect of time scheduling is called a session. We assume that a company sets the objective of selecting the popular programmes in $k$ parallel sessions among different channels so as to make its commercial advertisement reach the maximum number of viewers, that is, a company selects $k$ suitable programme slots simultaneously for advertisement. The aim of the paper is, therefore, to {help} the companies to select the programme slots, which are mutually exclusive with respect to the time schedule of telecasting time, in such a way that the total number of viewers of the selected programme in $k$ parallel slots rises to the optimum level. It is shown that the solution of this problem is obtained by solving the maximum weight $k$-colouring problem on an interval {graph}. An algorithm is designed to solve this just-in-time optimization problem using $O(kMn^2)$ time, where $n$ and $M$ represent the total number of programmes of all channels and the upper bound of the viewers of all programmes of all channels respectively. The problem considered in this paper is a daily life problem which is modeled by $k$-colouring problem on interval graph.