Dang Luo, Muffarah Ambreen, Assad Latif and Xiaolei Wang
Electricity plays an important role in the economic condition of any country. Nowadays, Pakistan is badly affected by shortage of electricity, which directly affected the economic…
Abstract
Purpose
Electricity plays an important role in the economic condition of any country. Nowadays, Pakistan is badly affected by shortage of electricity, which directly affected the economic growth of state. The purpose of this study is to propose an improved grey model DGPM(1,1,N) to forecast Pakistan's production of electricity, installed capacity and consumption.
Design/methodology/approach
To significantly simulate and predict accuracy, the discrete grey polynomial model DGPM(1,1,N) is improved with new information priority accumulation. The particle swarm optimization (PSO) algorithm is used for parameter optimization. The value of parameter is adjusted into improved grey model. By adjusting the parameter value in the model, the accuracy of prediction is enhanced.
Findings
The installed capacity of electricity needs more attention to improvement through implementation of effective polices, resolving major issues and funding scheme to fulfill the electricity demand of country. And improved DGPM(1,1,N) has better accuracy than original DGPM(1,1,N), DGM(1,1), nongrey models, linear regression and Holt–Winters methods.
Practical implications
This paper provides a practical and efficient improved grey method to predict the electricity production, consumption and installed capacity in Pakistan. This research and suggestion will help Pakistani government to formulate better policies to decrease the consumption of electricity and increase the installed capacity of electricity.
Originality/value
This paper not only improves the grey model with accumulation generation operator but also forecasts Pakistan's electricity production, installed capacity and consumption. It is a new idea to predict the installed capacity of electricity and the findings provide suggestions for the government to make policies.