Artificial Neural Network Based Optimum Scheduling and Management of Forecasting Municipal Solid Waste Generation - Case Study: Greater Noida in Uttar Pradesh (India)

• Published in: Journal of physics. Conference series 2020-04, Vol.1478 (1), p.12033
• Main Authors: Kumar, Sujit, Gaur, Aditya, Kamal, Neel, Pathak, Mohit, Shrinivas, Kumar, Singh, Priyanshu
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Datasets; Mathematical models; Mean square errors; Municipal solid waste; Municipal waste management; Neural networks; Parameters; Physics; Solid waste management; Time series
• ISSN: 1742-6588; 1742-6596
• DOI: 10.1088/1742-6596/1478/1/012033

Precise forecast of municipal strong waste era has a critical part in future arranging and squander management framework. The attributes of the created strong waste are distinctive at better places (region to district or nation to nation). The precise forecast of municipal solid waste (MSW) era turns into an essential errand in present day period. Its prediction requires accurate MSW data. The point of the present review is to outline the time series model for foreseeing month to month based strong waste production in Greater Noida city of Uttar Pradesh State (India) utilizing artificial neural network (ANN) with time series autoregressive method. The gathered municipal waste perceptions have been organized month to month from 2012 to 2016. The 60 months data set has been classified into 42 training data sets, 9 testing data sets and 9 validating data sets. An assortment of models of ANN has been examined by altering the number of hidden layer neurons. Ultimately, paramount enhanced architecture of neural network is established. The least value of performance parameters is validated in the proposed model as mean square error 0.0004, root mean square error 0.0203 and the high value of the coefficient of regression 0.8123. On the premise of these execution parameters it is reasoned that the ANN model provides precise prescient outcomes.