Microsoft Word - 164.docx CHEMICAL ENGINEERING TRANSACTIONS VOL. 56, 2017 A publication of The Italian Association of Chemical Engineering Online at www.aidic.it/cet Guest Editors: Jiří Jaromír Klemeš, Peng Yen Liew, Wai Shin Ho, Jeng Shiun Lim Copyright © 2017, AIDIC ServiziS.r.l., ISBN978-88-95608-47-1; ISSN 2283-9216 Optimal Location and Allocation for the Development of Oil Palm Eco-Industrial Town: Case Study in State of Johor Mohamad Firdza Shukerya,b, Haslenda Hashim*,a, Jeng Shiun Lima aProcess Systems Engineering Center (PROSPECT), Universiti Teknologi Malaysia, Johor, Malaysia bDepartment of Biological and Agricultural Engineering, Universiti Putra Malaysia, Selangor, Malaysia haslenda@cheme.utm.my In 2015, there were about 160 Mt of oil palm biomass produced by the oil palm industry nationwide. In this article, oil palm based eco-industrial town (EIT) is proposed to improve the profitability of oil palm industry while optimally utilise the availability of oil palm biomass and network infrastructure. The oil palm EIT is the integration of nine palm oil based industry and a community at the same location to promote energy and material sharing via industry symbiosis. The energy usage and waste generation are expected to reduce within the collaborated industries in the EIT. Due to the geographical location of existing palm oil mills in State of Johor, a crucial decision is required to choose a cost effective operating system and location for the development of oil palm EIT. A mixed integer linear programming (MILP) model is formulated to maximise economic performances, while selecting a cost effective operational system, either to employ centralised or decentralised system. Geographical information system (GIS) tool ArcGIS 10 was used to identify the location of existing crude palm oil mills, potential locations of oil palm EIT, and transportation network including roads and railways. Network analysis was performed to rank and calculate the most efficient network to transport oil biomass from oil palm plantation (surrounding existing palm oil mills) to the potential EIT locations via roads and railways. Generated results from network analysis were formulate as MILP model and optimised using General Algebraic Modelling System (GAMS). The optimisation results show that decentralised oil palm EIT system was the most profitable system with USD 241.54 M/y profit. The optimised model also suggested that the most efficient way to utilise abundant oil palm biomass are via crude palm oil mill, paper mill, livestock production, medium density fibre board mill, bio-diesel and bio-gas mill. The model shall assist the decision maker to identify system, location and the sub- industries in the EIT in order to promote sustainability in oil palm industry and bring benefits to the industry and community. 1. Introduction Malaysian oil palm industry has generated about 160 Mt of oil palm biomass annually (MPOB, 2015). These biomasses are included fresh fruit bunches (FFB), empty fruit bunches (EFB), oil palm frond (OPF), oil palm trunk (OPT), and palm oil mill effluent (POME). Improper management of oil palm biomasses lead to the resources loss, carbon dioxide emission, and environmental pollutions. In order to improve profitability and environment performance of oil palm industry, conceptual of oil palm eco-industrial town (EIT) is proposed. Integration of nine oil palm-based industries and a community at one location allows collaboration in raw materials, utilities, wastes, and energy usage. A sustainable and integrated bio-refinery concept for a palm oil mill was proposed by Ali et al. (2014). The concept targeted a zero emission system, which could produce higher value-added products from oil palm biomass such as bio-fertiliser, bio-char, biofuels, and biomaterials (Ng et al., 2014) presented a symbiotic bioenergy park that can fairly distribute cost and benefits among the collaborative industries. Most of these methodologies are exclusively designed to improve the crude palm oil mill as individual at one location and not consider centralising available biomass from others resources location. There are about 65 crude palm oil mills and 739,583 ha of oil palm plantations scattered in Johor, Malaysia. Upgrading the all existing crude palm oil mill into the oil palm EIT might be too costly since the oil palm EIT integrates nine oil palm based industry. Centralised resource could be more productive and cost efficient for development of oil palm EIT. Lim et al. DOI: 10.3303/CET1756237 Please cite this article as: Shukery M.F., Hashim H., Lim J.S., 2017, Optimal location and allocation for the development of oil palm eco- industrial town (eit): case study in state of johor, Chemical Engineering Transactions, 56, 1417-1422 DOI:10.3303/CET1756237 1417 (2013) developed a framework for the optimal design and planning of the product portfolio and processing route of an integrated, resource-efficient (IRE) rice mill complex. The objective function is to maximise the profitability of the rice mill by using the developed multi-period mathematical model. There is less of study on resource allocation for development of oil palm EIT. The main objective of this study is to develop a mathematical model that can select the optimal operating system for development of oil palm EIT, whether to choose centralised or decentralised system. In this study, a mixed integer linear programming (MILP) model is formulated to maximise economic performances, while choosing the most optimal operating system and location for the development of the oil palm EIT. GIS tool (ArcGIS 10) was used to plot the location of existing crude palm oil mills, potential locations of oil palm EIT, and transportation network including roads and railways. Network analysis was performed to rank and calculate the most efficient network to transport oil biomass from the existing palm oil mills to the potential EIT locations via roads and railways. Generated results from network analysis were then used in the developed MILP model and coded into General Algebraic Modelling System (GAMS) as an optimisation tool. GAMS version 23.5 was used to solve optimisation problems in this research due to the GAMS specially designed for modelling of linear, nonlinear and mixed-integer optimisation problems. Over the years, GAMS had been widely used to solve optimisation problem throughout the world and is proven useful for handling large and complex problems which may require many revisions to establish an accurate model. 2. Materials and method 2.1 Case study Kluang district of Johor was selected as case study due to the number of crude palm oil mills and vast oil palm plantation areas. Five conventional crude palm oil mills and two potential locations for development of oil palm EIT are taken as case study. The existing five mills are assumed to create decentralised system, while the two potential locations are assumed to create centralised system for the oil palm EIT. Plantation for each crude palm oil mill is assumed to be located within 5 km of radius. Each of oil palm EIT is assumed to have a palm oil mill, a community, a livestock production of cows, bio-fertiliser mill, bio-gas mill, bio-diesel mill, bio-pellet mill, medium-density fibre (mdf) board mill, and paper mill. All plants in the EIT are assumed to operate 20 h daily for 310 d annually. 2.2 Superstructure Development Figure 1 presents the simplified superstructure of the oil palm EIT, as described in the case study. The model comprises of five main sets, which are set a- types of input, set y-types of system, set n- types of industry output, set s- market, and set p for treatment facility. There are two types of external input resources to the industry or community, which are the main material i and utility f. Based on the main material that is processed by the industry or community, each industry in EIT can produce various products i.e. main product m, utility u, and by- product l. Main products and utilities generated by EIT can be used among the industries in EIT or sale to the market, s. The generated by-product can either be utilised by industries in the EIT or send the treatment facility. Set- v.r Set- v.n Set- v.a i1 UTYSALEus EXMATir Subset i (a) Set p set s Market PROSALEms PPROrm PUTYru PROUSEmr UTYUSEur BIOUSElr Subset f (a) Subset m (n) Subset u (n) Subset l (n) Treatment facility WASTElp i2 iN f1 f2 fN v1 v3 vN v2 m1 m2 mN u1 u2 uN l1 l2 lN v1 v2 va (v) vb (v) EXUTYfr PUTYru PBIOrl Material Utility Legend Figure 1: Simplified superstructure diagram of oil palm EIT 1418 2.3 Model Formulation In this section, definition of key terms in this study are first made, followed by defining the sets, variable and parameters, prior to the formulation of objective function and constraints (see Table 1). Table 1: List of indices, parameters and variables List of indices a Index for external resource v Index for location va(v) Location for decentralised system vb(v) Location for centralised system n Index for industry output s Index for market p Index for treatment facility List of parameters RESa,vs Amount of available external resource (t/y) CPCOSTv,r Amount of capital cost of industry y (USD/y) CNVSyvvrvsn Conversion factor of industry y to produce internal resource n DSTNCiavvsr Travel distance of external resource i to industry y (km) List of continuous variables EITPROFIT Total revenue of oil palm EIT (USD/y) SALES Revenue of sales (USD/y) UTYCOST Expenses for utility (USD/y) CAPCOST Expenses for capital cost (USD/y) MATCOST Expenses for feed cost (USD/y) TRANSCOST Expenses for transportation (USD/y) TREATCOST Expenses for waste treatment (USD/y) RMATvr Amount of main material at industry y (unit/y) PPROv,r,vs,m Amount of product n produced from industry y (unit/y) PBIOyv,r,vs,l Amount of biomass n produced from industry (unit/y) WASTEvs,l,p Amount of biomass n sent to treatment facility p (unit/y) PUTYv,r,vs,u Amount of utility n produced from industry y (unit/y) List of binary variables YA Selection for decentralised system YB Selection for centralised system XVBvb Selection for centralised location 2.3.1 Objective function The objective function of this model is to maximise the total revenue of the oil palm EIT, which represent by the variable EITPROFIT. Variable EITPROFIT consists of SALES, MATCOST, CAPCOST, UTYCOST, TRANSCOST, and TREATCOST variables as shown in Eq(1) to Eq(7). 𝑃𝑅𝑂𝐹𝐼𝑇𝑆 = 𝑆𝐴𝐿𝐸𝑆 − 𝑀𝐴𝑇𝐶𝑂𝑆𝑇 − 𝐶𝐴𝑃𝐶𝑂𝑆𝑇 − 𝑈𝑇𝑌𝐶𝑂𝑆𝑇 − 𝑇𝑅𝐴𝑁𝑆𝐶𝑂𝑆𝑇 − 𝑇𝑅𝐸𝐴𝑇𝐶𝑂𝑆𝑇 (1) 𝑆𝐴𝐿𝐸𝑆 = ∑ 𝑃𝑃𝑅𝑂𝑣,𝑟,𝑣𝑠,𝑚 × 𝐼𝑁𝑃𝑅𝑚 𝑣,𝑟,𝑣𝑠,𝑚 + ∑ 𝑃𝑈𝑇𝑌𝑣,𝑟,𝑣𝑠,𝑢 × 𝐼𝑁𝑃𝑅𝑢 𝑣,𝑟,𝑣𝑠,𝑢 (2) 𝑀𝐴𝑇𝐶𝑂𝑆𝑇 = ∑ 𝐸𝑋𝑀𝐴𝑇𝑖,𝑣𝑠,𝑣,𝑟 × 𝐸𝑋𝑃𝑅𝑖𝑖,𝑣𝑠,𝑣,𝑟 + ∑ 𝑃𝑅𝑂𝑈𝑆𝐸𝑣𝑠,𝑚,𝑣,𝑟 × 𝐼𝑁𝑃𝑅𝑚𝑣𝑠,𝑚,𝑣,𝑟 + ∑ 𝐵𝐼𝑂𝑈𝑆𝐸𝑣𝑠,𝑙,𝑣,𝑟 ×𝑣𝑠.𝑙.𝑣𝑟 𝐼𝑁𝑃𝑅𝑙 (3) 𝐶𝐴𝑃𝐶𝑂𝑆𝑇 = ∑ 𝐶𝑃𝐶𝑂𝑆𝑇𝑣𝑎,𝑟 × 𝑌𝐴 𝑣𝑎,𝑟 + ∑ 𝐶𝑃𝐶𝑂𝑆𝑇𝑣𝑏,𝑟 × 𝑋𝑉𝐵 𝑣𝑏,𝑟 (4) 𝑈𝑇𝑌𝐶𝑂𝑆𝑇 = ∑ 𝐸𝑋𝑈𝑇𝑖,𝑦 × 𝐸𝑋𝐶𝑂𝑆𝑇𝑓 𝑓,𝑦 + ∑ 𝑈𝑇𝑈𝑆𝐸𝑛,𝑦 × 𝐼𝑁𝐶𝑂𝑆𝑇𝑦 𝑛,𝑦 (5) 𝑇𝑅𝐴𝑁𝑆𝐶𝑂𝑆𝑇 = ∑ 𝐸𝑋𝑀𝐴𝑇𝑎,𝑣𝑠,𝑣,𝑟 × 𝑅 𝑎,𝑣𝑠 × 𝐷𝐼𝑆𝑇𝐴𝑁𝐶𝐸𝑎,𝑣𝑠,𝑣,𝑟 (6) 𝑇𝑅𝐸𝐴𝑇𝐶𝑂𝑆𝑇 = ∑ 𝑊𝐴𝑆𝑇𝐸𝑣𝑠,𝑙,,𝑝 × 𝑇𝑅𝐶𝑂𝑆𝑇𝑙 𝑣𝑠,𝑙,,𝑝 (7) 1419 2.3.2 Constrains The maximisation of the objective function of EITPROFIT is subjected to the following constraints: Eq(8) is formulated to ensure the oil palm EIT has enough supply of raw materials and utilities from external resources. The raw material at industry in EIT may come from external and internal resources as shown in Eq(9). Eq(10) indicated the products, utilities, and by-products produced by the industries and community in EIT. The summation of products, utilities and by-products from all industries and community in EIT are expressed in Eq(11). Products and utilities generated by EIT can be used by the industries or sale to market as bound by Eq(12). Eq(13) to Eq(18) is formulated to select of operating system and location for development of oil palm EIT. 𝑅𝐸𝑆𝑖,𝑣𝑠 ≥ ∑ 𝐸𝑋𝑀𝐴𝑇𝑖,𝑣𝑠,𝑣,𝑟 𝑣,𝑟 ∀ 𝑖 ∀ 𝑣𝑠 (8) 𝑅𝑀𝐴𝑇𝑣,𝑟 = ∑ 𝐸𝑋𝑀𝐴𝑇𝑖,𝑣𝑠,𝑣,𝑟 + 𝑖,𝑣𝑠 ∑ 𝑃𝑅𝑂𝑈𝑆𝐸𝑣𝑠,𝑚,𝑣,𝑟 + 𝑣𝑠,𝑚 ∑ 𝐵𝐼𝑂𝑈𝑆𝐸𝑣𝑠,𝑙,𝑣,𝑟 𝑣𝑠,𝑙 ∀ 𝑟 ∀ 𝑟 (9) 𝑅𝑀𝐴𝑇𝑣,𝑟 × 𝐶𝑁𝑉𝑆𝑣,𝑟,𝑣𝑠,𝑚, = 𝑃𝑃𝑅𝑂𝑣,𝑟,𝑣𝑠,𝑚 ∀ 𝑣∀ 𝑟 ∀ 𝑣𝑠 ∀ 𝑛 (10) ∑ 𝑃𝑃𝑅𝑂𝑣,𝑟,𝑣𝑠,𝑚 = 𝑣,𝑟 𝑃𝑅𝑂𝑣𝑠,𝑚 ∀ 𝑣𝑠 ∀ 𝑣𝑚 (11) 𝑃𝑅𝑂𝑣𝑠,𝑚 = ∑ 𝑃𝑅𝑂𝑈𝑆𝐸𝑣𝑠,𝑚,𝑣,𝑟 + ∑ 𝑃𝑅𝑂𝑆𝐴𝐿𝐸𝑣𝑠,𝑚,𝑠 ∀ 𝑣𝑠 𝑠𝑣,𝑟 ∀ 𝑚 (12) 𝑅𝑀𝐴𝑇𝑣𝑎,𝑟 ≤ ∑ 𝑅𝐸𝑆𝑖,𝑣𝑠 × 𝑌𝐴 𝑖,𝑣𝑠 ∀ 𝑣𝑎 ∀ 𝑟 (13) 𝑅𝑀𝐴𝑇𝑣𝑏,𝑟 ≤ ∑ 𝑅𝐸𝑆𝑖,𝑣𝑠 × 𝑌𝐵 𝑖,𝑣𝑠 ∀ 𝑣𝑏 ∀ 𝑟 (14) 𝑌𝐴 + 𝑌𝐵 ≤ 1 (15) 𝑅𝑀𝐴𝑇𝑣𝑏,𝑟 ≤ 𝐶𝑃𝑃𝐶𝑂𝑆𝑇𝑣𝑏,𝑟 × 𝑋𝑉𝐵𝑣𝑏 ∀ 𝑣𝑏 ∀ 𝑟 (16) 𝑅𝑀𝐴𝑇𝑟𝑑 ≤ ∑ 𝑅𝐸𝑆𝑖 × 𝑌𝐵 𝑖 ∀ 𝑟𝑑 (17) ∑ 𝑋𝑉𝐵𝑣𝑏 ≤ 1 𝑣𝑏 (18) 2.4 Data Input Table 2 illustrated the material and utility ratio (for 1 t basis of main material) at respective industry in the oil palm EIT. For example, if 1 t of fresh fruit bunches feed in at crude palm oil mill, the mill required about 0.0006 t of diesel and 1.26 t of water as material or utility input; and produces about 0.2 t of crude palm oil (key-product), 0.06 t of palm kernel, 0.23 t of empty fruit bunches, 0.65 t of pome and 17 kWh of electricity as material or utility output. Table 3 showed the information on existing palm oil mills at Kluang district. Transportation cost was taken from real field cost data which was USD 0.6 / km-t. 2.5 GIS data The ArcMAp 10 was used as GIS tool to plot location of existing crude palm oil mills, potential location for centralised system, and transportation link of roads and railways in the State of Johor. There were 65 crude palm oil mills and 23 potential locations were considered. Figure 2 showed the plotted GIS map for location of crude palm oil mills, potential centralised system location for EIT development and transportation link via roads and railways. Network analysis tool was used to calculate the most cost effective link between resource and the potential centralised system location. Table 4 summarised the distances via roads that produced from network analysis tool. 1420 Table 2: Material and utility ratio for industry in the oil palm EIT Input Industry Output Source Material Amount Material Amount FFBa 1 t Crude palm oil mill CPOb 0.23 t (Patthanaissaranukool et al., 2013) Diesel 0.0006 t PK 0.06 t Electricity 14.5 kWh POME 0.65 t Water 1.26 t EFB 0.23 t Bioelectricity 17 kWh EFBa /OPTa 1 t Bio-fertiliser mill Bio-fertiliserb 0.4 t (Yoshizaki et al. 2013) Diesel 0.0015 t Water 1 t OPFa 1 t Paper mill Paperb 0.5 t (Singh et al. 2013) Electricity 17 kWh POMEaDunga 1 t Bio-gas mill Bio-electricityb 39.6 kWh (Yoshizaki et al. 2013) Electricity 9 kWh Sludge 0.06 t amain material bkey product Table 3: Existing palm oil mills and potential centralised EIT in Kluang District, Johor Operating system Palm oil mill Plant size (t/h) Annual capacity (t/y) Annual capital and maintenances cost (USD/y) Plantation area (ha) Decentralised system KKS Bukit Lawiang 40 248,000 388,542 6,260.87 KKS Belitong 54 334,800 465,198 8,452.17 KKS Ulu Remis 60 372,000 495,556 9,347.83 KKS Southern Malay 40 248,000 388,542 6,260.87 KKS Air Hitam 40 248,000 388,542 6,260.86 Centralised system EIT A (Ayer Hitam) 200 1,240,000 1,020,516 - EIT B (Kbr Cina) 200 1,240,000 1,020,516 - Table 4: Distances (km) between oil palm biomass resources to the oil palm EIT Location KKS Bukit Lawiang KKS Belitong KKSUlu Remis KKS Southern Malay KKS Air Itam EIT A (Ayer Hitam) EIT B (KbrCina) Bukit Lawiang 5 8 47.59 33.45 48.43 32.77 13.78 Belitong 8 5 50.65 36.94 45.52 39.45 19.40 Ulu Remis 47.59 50.65 5 16.13 36.89 32.34 34 Southern Malay 33.45 36.94 16.13 5 20.89 16.74 20.06 Air Hitam 48.43 45.52 36.89 20.89 5 13.26 32.85 Figure 2: GIS map of crude palm oil mills location, centralised location and transportation link in State of Johor 1421 3. Results and Discussion The MILP model described in Section 2.3 was coded in GAMS with the objective to maximise the total annual oil EIT profit. The result indicates that the optimal system for developing oil palm EIT is by using decentralised system. Based on the objective function, the annual oil palm EIT profit of the system was revealed as USD 241.54 M (see Table 5). The highest expense was material cost with USD 147.63 M. Followed by capital and maintenances cost with USD 8.22 M, transportation cost with USD 7.90 M, and utility cost with USD 3.84 M. The oil palm EIT can reduce the cost of external utilities by utilising bio-diesel produced by the internal industries in the EIT. The lowest cost was the expense for treatment cost with USD 0.17 M. This is due to the most of the produced oil palm biomasses were used by the industries in EIT. The model has selected the decentralised system is mainly due by the high transportation cost required by the centralised system. The optimised model also suggested that the most efficient way to utilise abundant oil palm biomass are via crude palm oil mill, paper mill, livestock production, medium density fibre board mill, and bio-gas mill. Table 5: Optimal result of oil pam EIT Description Variable Value (M USD/y) Total revenue of oil palm EIT EITPROFIT 241.54 Revenue of sales SALES 409.29 Expenses for material cost MATCOST 147.63 Expenses for capital maintenances cost CAPCOST 8.22 Expenses for utility UTYCOST 3.84 Expenses for transportation cost TRANSCOST 7.90 Expenses for treatment cost TREATCOST 0.17 Selection of decentralised system YA 1 Selection of centralised system YB 0 4. Conclusion In this study, the model to maximise the economic performances of oil palm EIT by considering the logistic and availability of oil palm biomass, centralised and decentralised operational system and integrated biomass pathway for production of value added products has been successfully developed. It was found that the abundant supplies of palm oil biomass offer a wide range of investment prospects in the palm oil downstream sectors i.e. multi bio-products and bio-energy and fighting poverty through community development. The model is useful to assist the town planner and decision maker to select operational system, location, and sub- industries for development of the oil palm EIT. Acknowledgments The authors gratefully acknowledge the funding support for this work provided by Ministry of Education, Malaysia (MOHE) grant no. 2546.14H46, and Universiti Teknologi Malaysia (UTM). Reference Mohd Ali A.A., Othman M.R., Shirai Y., Hassan M.A., 2015, Sustainable and Integrated Palm Oil Biorefinery Concept with Value-Addition of Biomass, Journal of Cleaner Production 91, 96-99. Lim J.S., Abdul Manan Z., Wan Alwi S.R., Hashim H., 2013, A Multi-Period Model for Optimal Planning of an Integrated, Resource-Efficient Rice Mill, Computers ans Chemical Engineering 52 (13), 77–89. MPOB (Malaysia Palm Oil Board), 2015, Overview of the Malaysian Oil Palm Industry 2014 accessed 09.06.2015. Ng R.T.L., Ng D.K.S., Tan R.R., 2015, Optimal Planning, Design and Synthesis of Symbiotic Bioenergy Parks, Journal of Cleaner Production 87, 291–302. Patthanaissaranukool W., Polprasert C., Englande A.J., 2013, Potential Reduction of Carbon Emissions from Crude Palm Oil Production Based on Energy and Carbon Balances, Applied Energy 102 (C), 710–717. Singh P., Sulaiman O., Hashim R., Peng L.C., Singh R.P., 2013, Using Biomass Residues from Oil Palm Industry As a Raw Material for Pulp and Paper Industry: Potential Benefits and Threat to the Environment, Environment, Development and Sustainability, 15 (2), 367–383. Yoshizaki T., Shirai Y., Hassan M.A., Baharuddin A.S., Raja Abdullah N.M., Sulaiman S., Busu Z., 2013, Improved Economic Viability of Integrated Biogas Energy and Compost Production for Sustainable Palm Oil Mill Management, Journal of Cleaner Production 44, 1–7. 1422