CHEMICAL ENGINEERING TRANSACTIONS VOL. 76, 2019 A publication of The Italian Association of Chemical Engineering Online at www.aidic.it/cet Guest Editors: Petar S. Varbanov, Timothy G. Walmsley, Jiří J. Klemeš, Panos Seferlis Copyright © 2019, AIDIC Servizi S.r.l. ISBN 978-88-95608-73-0; ISSN 2283-9216 Circular Sustainability Optimisation Model for Diverse Oil Crops Feedstock System via Element Targeting Approach Chun Hsion Lima,*, William Wen Zhe Chuena, Jian Qi Fooa, Tee Jun Tana, Bing Shen Howb, Wendy Pei Qin Ngc, Hon Loong Lamd aUniversiti Tunku Abdul Rahman, Department of Chemical Engineering, Jalan Sungai Long, Bandar Sungai Long, Cheras, 43000 Kajang, Selangor, Malaysia bSwinburne University of Technology Sarawak Campus, Chemical Engineering Department, Faculty of Engineering, Computing and Science, Jalan Simpang Tiga, 93350 Kuching, Sarawak, Malaysia cCurtin University Malaysia, Department of Chemical Engineering, CDT 250, 98009 Miri, Sarawak, Malaysia dThe University of Nottingham Malaysia Campus, Department of Chemical Engineering, 43500 Jalan Broga, Semenyih, Selangor, Malaysia limchunhsion@yahoo.com Need of oil crops for bio-diesel production has been increasing due to its sustainable characteristic as compared to fossil fuel. Nonetheless, effort to ensure sustainable practise along bio-diesel production is essential to prevent back-fire from sustainable development. Each oil crop production and plantation activities are unique, leading to various in sustainability profile. In such context, systematic process optimisation for oil crop utilisation and selection is critical to ensure highest overall sustainable performance in the diverse oil crop system. This research adapts concept of element targeting in multi biomass selection to determine best diverse oil crop utilisation strategic. Various sustainable indexes are considered such as land used, carbon footprint, deforestation, water usage and fertiliser usage are considered. Cost calculation including transportation and production cost is incorporated in the study, while concept of circular economy is incorporated into the model by considering circular carbon element within the system. Demonstration case study is discussed to evaluate the performance of the model. The result shows that rapeseed oil is favourable in highest sustainability case, soybean provide lower cost and sunflower promote less carbon release. 1. Introduction There is an increasing demand for bio-diesel due to its sustainability characteristic as a renewable resource compared to the traditional fossil fuel. However, there have been various debates on the sustainability of oil crop plantations, such as deforestation of oil palm plantation in South East Asia regions. This has led to initiation of activities to ban palm oil utilisation in some Europe countries, disregard of the advantages of oil palm sustainable performance. The initiation has resulting major economy impact to the palm oil exporting countries in South East Asia. With the increasing demand for bio-diesel, diversification of bio-diesel source from various crops plantation is important to ensure sustainable, secure and consistent supply. Apart from that, different oil crops have dissimilar plantation activities and practises, leading to different sustainability profile in land usage, carbon footprint, deforestation, water usage and fertiliser usage. Thus, diversifying and mixing different oil crops in bio- diesel utilisation may help to improve overall sustainability profile for the industry. In December 2015, Paris Agreement was signed by 195 nations to reduce the global greenhouse gas emissions in order to keep global temperatures “well below 2 °C above preindustrial levels and pursuing efforts to limit the temperature increase to 1.5 °C (United Nations Framework Convention on Climate Change, 2015). Circular economy is a promising concept to evaluate and minimise carbon release from bio-diesel production based on regenerative and restorative of carbon footprint in plantation. High carbon release materials are problematic because removing carbon from environment required huge amount of investment and not sustainable. Recently, the Dutch government’s allocated 150 million euros for the construction of industrial-scale biodigesters to capture and valorise methane emissions from cow manure, which is expected to dent the country’s agricultural DOI: 10.3303/CET1976186 Paper Received: 15/03/2019; Revised: 01/06/2019; Accepted: 03/06/2019 Please cite this article as: Lim C.H., Chuen W.W.Z., Foo J.Q., Tan T.J., How B.S., Ng W.P.Q., Lam H.L., 2019, Circular Sustainability Optimisation Model for Diverse Oil Crops Feedstock System via Element Targeting Approach, Chemical Engineering Transactions, 76, 1111- 1116 DOI:10.3303/CET1976186 1111 emissions (Boztas, 2016). Similarly, carbon released from crop processing, transportation, cultivation and carbon fixation from plantation is critical to evaluate the sustainability of bio-diesel production. Complexity of sustainable bio-diesel production has led to development of various optimisation methods, such as an input- output model was suggested for sustainable oil palm plantation development by Foong et al. (2019). In logistic, Li and Zhao (2018) researched on the optimisation of chemical transportation using linear programming model. Whereas swarm particle optimisation was used by Wu et al. (2018) to optimise chemical productions. Nonetheless, none of the researcher has explored into development of sustainable bio-diesel development via feed diversification. In this research, a systematic oil crop utilisation model for process supply chain optimisation is proposed to consider the sustainability profile, production cost and circular carbon element in bio-diesel utilisation system. From the inspiration of multi-biomass selection model in Lim et al. (2018), Element Targeting approach is used in this study to consider multiple sustainability profile of various oil crops including fertiliser usage, land usage, water usage, carbon footprint and deforestation. Based on the acceptable range of each sustainable profile, a diverse oil crops utilisation system for bio-diesel production can be designed with optimum sustainability, cost and circular carbon element performance. In this study, land and sea transportation cost of bio-diesel produced from various oil crop is considered to determine the minimum requirement for production cost. With the concept of circular economy, carbon emission is determined based on carbon release and fixation from plantation activities-to-processing-to-product to identify circular sustainability of the system in terms of circular carbon element. With consideration of three aspects, i) oil crop sustainability profile, ii) production cost, and iii) circular carbon element, the proposed optimisation model can be used in oil crops management and development strategy to create a diverse bio-diesel utilisation system. 2. Methodology In order to obtain a diverse oil crop system with optimum sustainable characteristic, a two-step optimisation process is adapted to achieve an optimum performance of the system in terms of sustainability profile, production cost and circular element. Thus, the model proposed in this study includes multiple objectives/criteria, m optimisation that considered various oil crops, i and their respective sustainability indexes, j. The model also considered the supply chain network for oil crop distribution from respective resources location, r to respective process plant, p. Firstly, the proposed model is used to optimise each three criteria, m individually by i) maximising the sustainability profile, ii) minimising the total cost and iii) minimising the circular carbon element emission. By using the extreme cases of each aspect obtained from the first step, the overall performance can be evaluated in the second stage by maximising the overall performance score. 2.1 Sustainability profile criteria Sustainability score, Sij for each oil crop, i is evaluated based on each sustainable aspect, j including land usage, carbon footprint, deforestation, water usage and fertiliser usage. The sustainability indexes, SIij are calculated based on Eq(1) by comparing the sustainability score of each sustainable aspects, j with the maximum, S_maxij and minimum S_minij value among the considered oil crops, i. For example, score of 100 % is assigned to oil palm in land usage aspect due to highest oil yield per unit land. In contrast, 0 % is assigned in deforestation due to highest deforestation activity. SIij = 100 % × Sij - S_Minij S_Max ij - S_Min ij ∀ i,j ∈ I,J (1) Sustainable indexes of oil crops mixture, SI_Mixij is determined based on the individual indexes, SIij of each oil crop 𝑖 and sustainability aspect j, and the respective utilisation ratio of each crops, Xi.as shown in Eq(2). The summation of utilisation ratio of each oil crops is always equal to one as per Eq(3). To ensure the proposed mixture of oil crop utilisation fulfilled the requirement of each sustainability aspect, j, a constraint is introduced such that the sustainability indexes of the proposed oil crop utilisation is higher than the minimum sustainability profile, SI_minj requirement as shown in Eq(4). An overall sustainability index, Overall_SI in introduced to evaluate the performance of the oil crop utilisation mixture as shown in E(5). This equation is used as objective function for first step optimisation to maximise sustainability profile criteria. Priority factor Wj for each sustainability aspect j is introduced to improve the model flexibility to handle different scenarios. For example, higher priority factor value for carbon footprint index is set to ensure the result provides the best strategy to utilise mixture of bio-diesel to minimise carbon footprint. SI_Mix j = ∑[Xi × SIij] I i=1 ∀ j ∈ J (2) 1112 ∑ Xi = 1 I i=1 (3) SI_Mix j ≥ SI_Min j ∀ j ∈ J (4) Overall_SI = ∑ [ SI_Mix j × wj] J j=1 ∑ wj J j=1 (5) 2.2 Cost criteria Total production cost for each oil crop utilisation is also considered in this model. Transportation cost, Transport_cost is determined by using Eq(6), where Demand represents the demand of bio-diesel at bio-diesel mixing plant; dr i p represents truck travelled distance to deliver oil crops, i from resource r to plant p with road freight rate of RFR. Second term of the right-hand-side of Eq(6) covers transportation by sea. Sea freight rate, SFR is calculated based on Tanker World Scale Index as per Eq(7) (Muruganand, n.d.). Total production cost of bio-diesel, Production_cost is calculated based on production cost per unit of bio-diesel, Pi from each oil crops, i is calculated using Eq(8) and total cost of the system governed by Eq(9). This equation is used as the objective function to minimise the total cost, TC for cost criteria consideration. Transport_cost = ∑ (Demand × Xi× RFR × dr i p) R,I,P r,i,p=1 + ∑(Demand × Xi× SFR) I i=1 (6) SFR = Daily hire + Port Cost + Fuel cost + Canal dues Cargo Quantity (7) Production_cost = ∑( Demand × Xi × Pi) I i=1 (8) TC = Transport_cost + Production_cost (9) 2.3 Circular Carbon Element criteria The overall circular carbon element, Ci for each oil crops, i is evaluated based on carbon emission per land used, C_emiti and carbon fixation, C_fixi per land used as per Eq(10). Eq(11) shows calculation of total circular carbon element score, TCCE, where Yi is refers to oil yield per unit of land used for each corps, i to determine total land usage for plantation for net carbon emission calculation. This equation is used as objective function to minimise the net carbon emission of the system during circular carbon element criteria consideration. Ci= C_emiti - Carbon_fixi ∀ i ∈ I (10) TCCE = ∑ ( Demand × Xi Yi × Ci) I i=i (11) 2.4 Overall Performance Score of Oil Crop Mixture In order to obtain an overall performance score of the system based on the listed criteria, m, (sustainable index, cost, and circular carbon element), three extreme cases are generated to i) maximising Eq(5), ii) minimising Eq(9), and iii) minimising Eq(11). Similar to Eq(1), Eq(12) shows calculation of criteria index, CIm which is used to determine the level of satisfaction of each criteria score CSm with respect to the maximum criteria score, CS_maxm and minimum criteria score, CS_minm in the extreme cases. An overall performance of the system across all three criteria can be optimised as per Eq(13). Another priority factor, Fm is introduced to provide model flexibility to prioritise each criteria, m in different scenario or application. This equation is used as the objective function in second stage optimisation to maximise the overall performance, OP of the system. CIm = 100 % × CSm - CS_minm CS_max m - CS_min m ∀ m ∈ M (12) 1113 OP = ∑ [ CIm × Fm] M m=1 ∑ Fm M m=1 (13) 3. Case Study A demonstration case study is constructed to show the performance of the proposed model. Processing plant for bio-diesel mixing and utilisation is set to be in Germany, and the bio-diesel is obtained from various region, namely oil palm from Malaysia, rapeseed from Rostock, Germany, soybean from Verona, Italy and sunflower from Toulouse, France. The location of resources is determined based on the total production percentage in European Union (USDA, 2014). The proposed model is used to determine the optimum oil crop utilisation ratio based on sustainable index, cost and circular carbon element. Table 1 shows the data for sustainable index for each oil crops based on Eq(1). Five indexes were considered, namely deforestation, oil yield, fertiliser, carbon footprint, and water usage. The minimum score of each sustainable index of the oil crop mixture is set to be 40 % as per Eq(4). The individual oil crop deforestation percentage was determined based on ratio of total deforestation area and total plantation area from European Commission (2013) and Lima et al. (2011). The land usage data was extracted by considering the oil yield in gal/ acre extracted from Kurki et al. (2010) to estimate plantation land needed. Besides, the fertiliser emission of oil palm was extracted from Saswattecha et al. (2015), fertiliser emission of soybean was extracted from Zortea et al. (2018), while fertiliser emission of rapeseed and sunflower was extracted from Forleo et al. (2018). Moreover, the data for water usage of oil crops were extracted from Mekonnen and Hoekstra (2010) and sources for carbon footprint were obtained from Muñoz et al. (2014). The bio-diesel demand was estimated at 13.9 million tones and Table 2 shows the transportation distance (estimated from Google Map) from respective resource location to the process plant via truck with a fix transportation cost, RFR of $ 0.0807 /t.km. Noted that additional shipment transport of palm oil from Malaysia to German via sea freight was considered with the distance of 15,864 km at the rate, SFR of $ 0.0526 /kg bio - diesel. Production cost of bio-diesel from respective oil crops is also summarised in Table 2 with consideration of raw material cost and operating cost. Table 3 shows the calculation of carbon emission for each oil crops. Table 1: Sustainable indexes data for each oil crop Table 2: Land transportation distance of each bio-diesel to processing plant and its respective production cost Table 3: Carbon released and fixation by each oil crop Oil Palm Soybean Oil Rapeseed Oil Sunflower Seed Oil Score SI (%) Score SI (%) Score SI (%) Score SI (%) Land deforestation percentage (%) 23.10 0 14.90 36 0.40 100 3.10 88 Oil yield (gal oil/ acre) 610 100 46 0 122 14 98 9 Fertiliser emission (kg) 0.4035 63 0.0983 100 0.8830 4 0.9159 0 Carbon footprint(kg) 2,024 0 2,024 0 262 100 760 72 Water Usage (m3/ton) 1,098 100 2,271 54 2,145 48 3,366 0 Oil Palm Soybean Oil Rapeseed Oil Sunflower Seed Oil Land distance (km) 175 1,112 154 1,723 Production cost ($ /kg bio-diesel) 0.72 (Ong et al., 2012) 0.61 (Haas et al., 2006) 1.07 (El-Enin et. al., 2013) 1.06 (Adhikari & Illukpitiya, 2017) Oil Palm Soybean Oil Rapeseed Oil Sunflower Seed Oil Emission from cultivation (kg CO2/t of oil) 101.0 535.5 151.0 321.2 Emission from transportation (kg CO2/t of oil) 212.25 2.80 4.90 6.70 Emission from processing (kg CO2/t of oil) 51.00 293.30 450.55 140.00 Yield (t/ha/y) 4.5 2.2 3.1 2.5 Total emission (kg CO2/t/ha) 1,639.13 1,829.52 1,880.00 1,169.75 Carbon fixation (kg CO2/ha/y) -2,497.5 -2,098.0 -2,449.8 -2,084.0 Net emission, Ci (kg CO2/ha/y) -858.37 -268.48 -569.80 -914.25 1114 The model was solved using General Algebraic Modelling System (GAMS). Result from the first stage optimisation of each extreme cases (maximising Overall_SI, minimising TC and minimising TCCE) are tabulated in Table 4. All priority factors were set to be one to reflect equal importance of each sustainable index. In each case, optimum oil crops usage ratio, Xi were suggested for difference objective functions. The result shows approximately 46 % of bio-diesel from rapeseed oil, 43 % from oil palm and 11 % from soybean should be used in order to obtain maximum sustainable rating at 51.16 %. In other case, soybean oil crop utilisation should be prioritised in order to obtain lowest total cost. Likewise, sunflower should be utilised more to achieve minimum circular carbon element or carbon release. Based on the first stage optimisation, second stage optimisation was conducted to obtain the highest overall sustainable performance score. Priority factors, Fm were assumed to be one for equal priority of each aspect where the result shows an optimum overall performance of 57.2 %. The oil crop utilisation ratio for the suggested optimum result was found to be identical to the case of maximising Overall_SI in Table 4. In this optimum result, the sustainability criteria was achieved 100 %, cost criteria was achieved only 54 % and circular carbon element criteria was achieved the lowest at 18 % as show in Table 5. The individual sustainability index also tabulated in Table 5. Table 4: Result after optimising one of the aspects (1st step optimisation) Table 5: Each criteria fulfilment and mixture sustainability indexes for optimum overall performance case The mixture sustainable index SI_Mixj was calculated using Eq (2). The result shows fertiliser index fulfilled the minimum requirement of 40 % while water usage index achieved highest score at 71 %. Based on the result, the optimum oil crops ratio is favorable towards rapeseed oil and palm oil due to both oil crops have relatively higher sustainability score shown in Table 1. Thus, increase utilisation of both oil crops will provides better overall sustainable index performance. In contrast, soybean oil has lower utilisation fraction due to lack of performance in carbon footprint and land usage. This shows a significant improvement opportunity for soybean stakeholders to enhance overall carbon footprint and efficiency of land usage. Similarly, land and water usage should also be a concern in sunflower plantation to compete with other oil crop in term of sustainability. However, in the event where cost is the priority and value of 2 is assigned in cost criteria factor, new optimum solution was obtained. Utilisation of bio-diesel from oil palm was suggested as 0.2515, rapeseed as 0.3596, soybean as 0.3888, sunflower seed as 0, while the overall performance of system increases to 66.24 %. Similarly, when assigned circular carbon element as priority, the optimum crop utilisation fraction changes to oil palm as 0.3116, rapeseed as 0, soybean as 0.2047, sunflower seed as 0.4837, while the overall performance of system was 52.64 %. With the flexibility of the proposed model, utilisation strategic of each oil crop for bio-diesel production can be generated based on the priority of each sustainability index and criteria. Ultimately, this model can be used as a guideline to produce a diverse agriculture development, such as to strategise amount of plantation of each oil crop based on the proposed ratio to achieve better sustainable practises in bio-diesel production. 4. Conclusions In conclusion, this research proposed an optimisation model to determine multi oil crop utilisation in bio-diesel production with consideration of sustainability index, cost and circular carbon element. A demonstration case study was illustrated in this paper to determine oil crop utilisation strategy in bio-diesel production. Priority factor was used to enhance model flexibility for various scenarios, such as providing strategy to focus on cost saving CIm MAX Overall_SI MIN TC MIN TCCE Sustainable Rating of Mixture (%) 51.16 47.10 39.56 Total Cost ($ m) 12,659.7 11,920.8 13,515.5 Circular carbon element (t CO2/ t of oil) -2,493,590 -2,240,940 -3,642,950 Xi Oil Palm 0.4273 0.2515 0.2534 Rapeseed 0.4589 0.3596 0.0000 Soybean 0.1138 0.3888 0.2412 Sunflower 0.0000 0.0000 0.5054 MAX Overall_SI MIN TC MIN TCCE CIm 100 % 54 % 18 % Fertiliser (%) Carbon Footprint (%) Land usage (%) Deforestation (%) Water usage (%) SI_Mixi 40.00 45.89 48.92 50.00 71.00 1115 instead of sustainability profile. This provides flexibility to accommodate different stakeholders and their respective optimisation intention. Future improvement of model is possible by considering fluctuation of process performance based on differences in oil crop and bio-diesel properties. In addition, consideration of food supply chain in parallel with energy supply chain also will provide insightful oil crop plantation and development strategy. Acknowledgments The financial support by Universiti Tunku Abdul Rahman Research Fund: IPSR/RMC/UTARRF/2018-C2/L05. References Adhikari, S., and Illukpitiya, P., 2017, Comparative economic analysis of on-farm biodiesel production accessed 3.03.2019 Boztas, S., 2016, Poo power: Dutch dairy industry launches €150m biogas project, accessed 28.03.2019 El-Enin, S.A., Attia, N.K., El-Ibiari, N.N., El-Diwani, G.I., El-Khatib, K.M., 2013, In-situ transesterification of rapeseed and cost indicators for biodiesel production, Renewable and Sustainable Energy Reviews,18, 471- 477. European Commission, 2013, The impact of EU consumption on deforestation: Comprehensive analysis of the impact of EU consumption on deforestation, European Union, 2–348. Foong, S.Z.Y., Goh, C.K.M., Supramaniam, C.V., Ng, D.K.S., 2019, Input-output optimisation model for sustainable oil palm plantation development, Sustainable Production and Consumption, 17, 31-46. Forleo, M.B., Palmieri, N., Suardi, A., Coaloa, D., Pari, L., 2018, The eco-efficiency of rapeseed and sunflower cultivation in Italy. Joining environmental and economic assessment, Journal of Cleaner Production, 172, 3138-3153. Muñoz, I., Schmidt, J.H., Dalgaard, R., 2014, Comparative life cycle assessment of five different vegetable oils, Proceeding of the 9th International Conference on Life Cycle Assessment in the Agri-Food Sector, 886-894. Haas, M. J., Mcaloon, A. J., Yee, W. C., Foglia, T. A., 2006, A process model to estimate biodiesel production costs, Bioresource Technology, 97, 671-678. Kurki, A., Hill, A., Morris, M., 2010, Biodiesel: The sustainability dimensions, NCAT accessed 15.03.2019 Li, Y., Zhao, N., 2018, Hazardous Chemical Transportation Inputs and Optimization Based on Linear Programming Model, Chemical Engineering Transactions, 71, 517-522. Lim, C.H., How, B.S., Ng, W.P.Q., Leong, W.D., Ngan, S.L., Lam, H.L., 2018, Debottlenecking biomass supply chain resources deficiency via element targeting approach, Chemical Engineering Transactions, 70, 2155- 2160. Lima, M., Skutsch, M., Costa, G.D.M., 2011, Deforestation and the Social Impacts of Soy for Biodiesel: Perspectives of Farmers in the South Brazilian Amazon. Ecology and Society, 16(4). Muruganand, D., n.d, World Scale Basic accessed 10.03.2019 Mekonnen, M.M., Hoekstra, A.Y., 2011, The green, blue and grey water footprint of crops and derived crop products, Hydrology and Earth System Sciences, 15, 1577-1600. Ong, H. C., Mahlia, T. M. I., Masjuki, H. H., Honnery, D., 2012, Life cycle cost and sensitivity analysis of palm biodiesel production, Fuel, 98, 131–139. Saswattecha, K., Kroeze, C., Jawjit, W., Hein, L., 2015, Assessing the environmental impact of palm oil produced in Thailand, Journal of Cleaner Production, 100,150-169. United Nations Framework Convention on Climate Change, 2015, Adoption of Paris agreement accessed 15.03.2019. USDA, 2014, Europe- Crop Production Maps, United States Department of Agriculture accessed 13.03.2019. Wu, D., Wang, L., Ying, Y., 2018, Multi-step optimization of chemical production workshop based on improved particle swarm optimization, Chemical Engineering Transactions, 71, 421-426. Zortea, R.B., Maciel, V.G., Passuello, A., 2018, Sustainability assessment of soybean production in Southern Brazil: A life cycle approach, Sustainable Production and Consumption, 13,102-112. 1116