Design and development of an automatic grading machine for oil palm (Elaeis Guineensis Jacq.) fresh fruits bunch (FFB) using non-destructive examination | |
| Author | Makky, Muhammad |
| Call Number | AIT Diss. no.AE-14-01 |
| Subject(s) | Oil palm--Indonesia Machinery--Automatic control Machine-tools--Design |
| Note | A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Engineering in Agricultural System and Engineering |
| Publisher | Asian Institute of Technology |
| Abstract | Oil palm industry play s a major role in Indonesian economy , provid ing employment for millions of workers and generate s national income as well as foreign - exchange from the export. However, problems deterred the industry to progress, primarily due to poor quality and productivity, as the consequences of improper quality controls of raw materials, i.e. the oil palm fresh fruits bunch (FFB). Since they are still handled by manual grading throug h visual inspection , their results are heavily influenced by subjectivity. In consequence, a n efficient and effective grading system, with better consistency and accuracy , while being cost effective , is required . The objective of the study is to develop an automated grading machine for oil palm FFB, which can perform grading and determine the FFB properties by means of machine vision. The developed machine prototype comprised mechanical system and control system . T he mechanical system comprised the chassis , an electric motor driven conveyor - belt, electric motor actuated separator gate, and an enclosed inspection chamber. The control system comprised a machine vision system and control system . Control system comprise d a single chip microcontroller (SCM) module with an in - system programing (ISP) capability; a relay board for actuating conveyor and separator, as well as lighting; and a computer, communicat ing with SCM via USB connection . The machine vision comprised a c amera, a computer, and an image processing program, to capture FFB images, analyze it , remove the background, extract features from images and perform calculation of FFB class, fraction, weight, oil content (OC) and free fatty acids (FFA) level. Eleven predictor variables extracted from the captured image are used to develop models to estimate FFB class, fraction, weight, OC and FFA. Using stepwise discriminant analysis with Ma ha lanobis distance, the class model successfully classified FFB into tw o classes, with R 2 , SEC and SEP as 0.946, 0.5105, and 0.5605 respectively. K - means clustering and squared Euclidean distance analysis are employed to model the FFB fractions, resulting R 2 , SEC and SEP as 0.8625, 1.042, and 0.555 respectively. Eight fractio n - wise models are developed to estimate the FFB weight using cubical regression fit, which result in R 2 , SEC and SEP as 0.9176, 1.194 and 1.8182 accordingly. For modeling the FFB oil content and free fatty acid, forward stepwise multiple linear regression analysis is performed for both models. The OC model correctly estimated the oil content in the FFB mesocarp with R 2 , SEC and SEP as 0.9453, 0.40585, and 0.458 respectively. F or the FFA model, the R 2 , SEC and SEP are obtained as 0.8959, 0.06583, and 0.0658 accordingly. The working capacity of the prototype is calculated as 8 , 850 kg/h, while on tests the actual work capacity averaged as 6 , 135.8 kg/h. In term of economic analysis , the initial investment for the machine is approximately $ 5000, while the operat i onal cost of the machine is calculated as 0. 0002 7 $ /kg . The break - even point of machine is obtained as 4,204,444.4 kg/yr. The benefit to cost ratio of 26 . 8 indicated that substituting human grader with the developed machine will give attractive economic benefits to the oil palm mills. |
| Year | 2014 |
| Type | Dissertation |
| School | School of Environment, Resources, and Development (SERD) |
| Department | Department of Food, Agriculture and Natural Resources (Former title: Department of Food Agriculture, and BioResources (DFAB)) |
| Academic Program/FoS | Agricultural and Food Engineering (AE) |
| Chairperson(s) | Soni, Peeyush; |
| Examination Committee(s) | Athapol Noomhorm;Manukid Parnichkun;Salokhe, Vilas M.;Shimizu, Hiroshi; |
| Scholarship Donor(s) | Government of Republic of Indonesia (DIKTI); |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2014 |