A comparative study between Modified Data Envelopment Analysis and Response Surface Methodology for optimisation of heterogeneous biodiesel production from waste cooking palm oil. (10th November 2016)

• Record Type: Journal Article
• Authors: Saeidi, Samrand; Jouybanpour, Parvin; Mirvakilli, Azadeh; Iranshahi, Davood; Klemeš, Jiří Jaromír
• Is Part Of: Journal of cleaner production. Volume 136:Part B(2016)
• Page Start: 23
• Page End: 30
• Publication Date: 2016-11-10
• Subjects: Biodiesel production -- Waste cooking palm oil -- Transesterification -- Modified Data Envelopment Analysis (MDEA) -- Data envelopment analysis (DEA) -- Neural Network (NN) -- Response Surface Methodology (RSM); Factory and trade waste -- Management -- Periodicals; Manufactures -- Environmental aspects -- Periodicals; Déchets industriels -- Gestion -- Périodiques; Usines -- Aspect de l'environnement -- Périodiques; 628.5
• DOI: 10.1016/j.jclepro.2016.06.192
• Languages: English
• ISSNs: 0959-6526

Abstract

Abstract: Biodiesel is a clean renewable fuel which is an alternative source of diesel fuel in compression ignition engines without any modification. According to previous research, the importance of biodiesel production through heterogeneous transesterification of waste cooking palm oil (WCPO) over Sr/ZrO2 catalyst has led to developing a new mathematical algorithm called Modified Data Envelopment Analysis (MDEA). MDEA, a hybrid of Data Envelopment Analysis (DEA) with Neural Network (NN), was proposed for experiment design of multi-response problems. It was validated with Response Surface Methodology (RSM), which is a statistical method. This method was developed to maximise Fatty Acid Methyl Ester (FAME) yield and five decision variables were considered. The optimum amount of methanol to oil molar ratio, catalyst loading, reaction temperature, reaction time on Ester yield, and free fatty acid (FFA) conversion were calculated via MDEA method. The obtained results showed that the derived optimal parameter-setting of the proposed method, MDEA, is more reliable and accurate than RSM. The errors of predicted Ester yields are 5% and 14% in MDEA and RSM. The calculated errors of conversions are 3% and 19% in MDEA and RSM. Highlights: Biodiesel importance via WCPO transesterification was led to develop MDEA. MDEA include DEA and ANN for experiment design of multi response problem. MDEA and RSM were introduced as mathematical and statistical method, respectively. A good agreement was found between experimental and MDEA results. MDEA was accurate and reliable in comparison with RSM.