Yandra Arkeman

Solving fuzzy multi-objective optimization using non-dominated sorting genetic algorithm II

This paper presents the stages for solving fuzzy multi-objective optimization problems using genetic algorithm approach. Before applying non-dominated sorting genetic algorithm II (NSGA II) techniques to obtain optimal solution, first multi-objective possibilistic (fuzzy) programming was converted into an equivalent auxiliary crisp model to form deterministic programming model. To determine the best solution from Pareto set, we implied feasibility degree of decision variable and satisfaction degree of decision maker. The best optimal solution is the intersection between α-feasibility degree and satisfaction degree of the decision makers that has the highest fuzzy membership degree. For numerical experiment, we used simple formulation in multi-objective fuzzy linear programming model with three maximum objective functions, three decision variables, and six constraints. The comparison of the results shows that our results are better for two objectives than that of compromising programming.

PEMODELAN STATISTICAL CONTROL DETECTION ADAPTIVE (SCDA) UNTUK MONITORING DAN PREDIKSI VOLUME PRODUKSI CRUDE PALM OIL (CPO) NASIONAL

Achievement of national palm oil industry as a producer and exporter of crude palm oil (CPO) in the world, it is now giving birth insecurity issues. This is because the growth of upstream and downstream industries of national palm oil that has not been balanced, which in turn encourages the national palm oil industry players to be oriented to the export of CPO which eliminates the added value in the country. On the other hand, though bring in foreign exchange for the country, but is prone commodity export orientation encountered a barriers problem in the international market. It is therefore important to provide a means of monitoring, prediction and assessment to facilitate the formulation of policies more about the marketing of national CPO industry. This research proposed the development of a model framework called adaptive threshold statistical control detection adaptive (SCDA) as a means of monitoring, prediction, and assessment of the movement of national CPO production volume. SCDA idea is to determine the dynamic threshold based mapping pattern historical data and predictions from the aspect of the frequency and trends. SCDA model adapted the techniques of statistical process control (SPC), while the values of the predictions generated from the simulation prediction model developed using the techniques of artificial neural network back propagation (ANN-BP) based on historical data of the national CPO production volume. The data used was the average volume of annual national CPO production period 1967 to 2015. The simulation results showed that the prediction model of national CPO production volume in 2016 until 2018 predicted were31.025 million, 32.214 million, and 34.504 million tons, respectively, while the values of maximum and minimum threshold that was formed in the model predictions SCDA for the period 2016-2018 each sequence were 33,322,065 and 29,246,547, respectively. As far as the literature search results, modeling SCDA has never been done in the research included for monitoring and prediction of national production volume of CPO. Therefore, research on the modeling of SCDA was contributing both to the development of knowledge about modeling as well as in the management of the national supply of CPO. Keywords: adaptive threshold, modelling, artificial neural network, palm oil

Business Process Reengineering of Sustainable Teak Forest at Agroforestry Industry (pp 169-183)

This paper examines the relationship between family control and dividend policy in Indonesia. There are three possible explanations for the relationship. The expropriation hypothesis predicts that family control has a negative impact on dividend payouts. Meanwhile the reputation hypothesis and the family income hypothesis predict that family control has a positive impact on dividend payouts. Using a panel data of Indonesian publicly listed firms in the period of 2003-2009, the results shows that family control has a significant negative impact on dividend payouts, dividend yields and likelyhood to pay dividends. The results control for other variables that may potentially affect dividend payments such as growth opportunity, debt, profitability, firm size and firm age. From agency theory perspective, the finding is consistent with the argument that family controlling shareholders prefer lower dividends, in order to preserve cash flows that they can potentially expropriate (the expropriation hypothesis). Keywords: Family Control, Dividends, Agency Theory

The Design of Net Energy Balance Optimization Model for Crude Palm Oil Production

Net energy balance (NEB) is the second important indicator following green house gases in developing a sustainable biodiesel industry. The extent of the production chain and various ways to reduce the use of fossil fuels, increase the complexity of finding an optimal NEB value of the industry. The main objective of this study was to design an NEB optimization model, which was supported by genetic algorithm (GA). The model was applied in a crude palm oil (CPO) industry that produces raw material for biodiesel which is located in North Sumatra province. The model was solved by using an optimization computer software package. The results showed that the NEB value was better than the previous one. The model was also able to provide biomass usage composition to achieve the optimal NEB value, and the unit processes that need to be improved.

Genetic algorithm based multi-objective optimization of wheat flour supply chain considering raw material substitution

The aim of this study was to develop multi-objective optimization model for wheat flour supply chain. The model was developed by considering raw material substitution with local flour. The local flour such as mocaf, tapioca, sweet potato, modified corn flour etc. can substitute a part or whole of wheat flour for wheat flour-based product application. However, raw material substitution can impact supply chain network, raw material supply policy, and product quality so that it is important to optimize supply chain for that case. In this work, we used mocaf as flour substitution for wheat flour in wheat flour mill. We developed multi-objective supply chain model that minimized total cost and maximized product quality. Genetic algorithm approach was used to solve the optimization problem. For numerical experiment, we used supply chain configuration consisting of three wheat suppliers, three mocaf suppliers, three wheat flour mills, four distribution centers, and two food factories.

Design of web-based information system with Green House Gas analysis for palm oil biodiesel agroindustry

The scarcity of fuel is one of the serious problems in Indonesia because it can disturb people daily activities. One way to prevent the increasing fuel usage is by providing information about biodiesel as an alternative energy source. This information is served in the form of web base information for palm oil biodiesel agroindustry with Green House Gas (GHG) analytical module. This research is aimed at providing information about data production and area of oil palm plantation, data production of biodiesel and GHG analysis module to perceive GHG emission. This information system has some main features including graph of data production, area of oil palm plantation, biodiesel data production, page editor, and shown results of GHG analysis. Using this information system, it can be decided whether an area has a high GHG emission or not. This system, therefore, can be used as a region reference for emission reduction.

The application of Fuzzy-Neuro approach for ERP system selection: Case study on an agro-industrial enterprise

Enterprise Resource Planning (ERP) adoption emphasizes business transformation that leads to change business processes in an effort to maximize profits and competitive advantage of the enterprise. Many companies were unsuccessful in implementing ERP system. Selection failure affected implementation failure. ERP system selection that misfit and ineffectively caused a major failure of ERP system adoption which is a critical investment, risky and expensive. ERP selection is a complex decision-making process and must be conducted carefully because of the important impacts. Many researchers have studied related to the approach used, but still little was associated with complex and standardized criteria. Most studies were to simplify the complex criteria, which often will eliminate the meaning of the standardized criteria. This study discusses the hybrid approach of Fuzzy - Neural Network (Fuzzy-Neuro) for the ERP selection with numerous and complex criteria. The criterions used were the characteristics and sub-characteristics that compatible with ISO25010, vendors and consultants, fit strategy, change management and cost. A case study was simulated in the agro industrial company that has some special characteristics. The results confirm the Fuzzy-Neuro approach can be used optimally even for ERP selection with many, complex and tiered standardized criteria.