Municipal Waste Generation Forecasting Based On Advanced Hidden Markov Model

Forecasting of municipal waste generation gives theoretical advice towaste treatment capacity design and thus critical for municipal sustainabledevelopment.Traditional methods use the idea of average and cannot measuredynamic fluctuations and dynamically trace peek accurately. Due to thefact that waste generation is highly affected by economics, cultures andenvironment, this paper presents advanced Hidden Markov Model basedon Gaussian Mixture Model. Choose small dataset Shanghai solid wastegeneration and big dataset American wastewater quantity from open sourceas case, respectively adopt state transform to calculate expected forecast orseek the most relevant historical pattern through posterior probability, anduse bootstrap to give the confidence interval. The results prove the newmethod is efficient and applicable.So far, we have already obtained foresting result of Shanghaimunicipal waste generation from Year2004to Year2010from theaforementioned improved GMM-HMM. Hidden Markov Model, the sameas other traditional time series projection methods, only focuses on historyinformation and doesn’t allow for the influence of future policy changes.So, the next step is using the results getting from aforementioned improvedGMM-HMM as the input source ‘waste generation’ to the SystemDynamic model of Shanghai municipal treatment waste system. Wastegeneration consists of waste recycling, illegal disposal and waste treatment.Waste treatment quantity forecasting can be utilized to help make decisionstowards future capacity design of waste treatment facilities. Build up different scenarios by changing recycling market price, waste treatmentprice and the degree of regulation towards illegal disposal. Finally weobtain Shanghai municipal waste treatment quantity from Year2004toYear2010in different scenarios. The sensitive analysis indicates thatincreasing waste treatment price is the most efficient method to cut backon the waste treatment quantity, and further help to alleviate the burden ofShanghai Municipal Waste Treatment System.