Carbon Footprint Calculation and Optimization Approach for CFOOD Project Piotr Milczarski Artur Hłobaż Paweł Maślanka Bartosz Zieliński Faculty of Physics and Faculty of Physics and Faculty of Physics and Faculty of Physics and Applied Informatics Applied Informatics Applied Informatics Applied Informatics University of Lodz University of Lodz University of Lodz University of Lodz Lodz, Poland Lodz, Poland Lodz, Poland Lodz, Poland piotr.milczarski@uni.lodz.pl artur.hlobaz@uni.lodz.pl pmaslan@uni.lodz.pl bartosz.zielinski@uni.lodz.pl Zofia Stawska Krzysztof Podlaski Piotr Kosiński Faculty of Physics and Applied Faculty of Physics and Applied Faculty of Physics and Applied Informatics Informatics Informatics University of Lodz University of Lodz University of Lodz Lodz, Poland Lodz, Poland Lodz, Poland zofia.stawska@uni.lodz.pl krzysztof.podlaski@uni.lodz.pl pkosinsk@uni.lodz.pl Abstract—In the paper, the study of carbon footprint The growing population also needs more food especially optimization process is shown in order to receive low-carbon processed food due to increased urbanization [4][5]. That products. A short description of the Carbon Footprint needs more supplies, raw materials and resources e.g. energy standards is provided. Basing on the conducting project ones. Hence, not only governments or institutions e.g. the EU CFOOD subsided by Polish R&D Agency the optimization commission impose higher demands on lowering the usage of boundaries are discussed and presented. In the paper, the the energy resources (coal, fuels, electricity and gas) but also methods of carbon footprint are discussed. Basing on life cycle companies e.g. the food processing ones. The companies in assessment (LCA) the model for carbon footprint is presented their food processes are interested in implementing low- and discussed. LCA is then implemented to assess carbon carbon technologies or solutions from economic reasons i.e. footprint at the manufacturing and transportation stages in the food processing industry. the less energy the cheaper product. It must be connected with the keeping-up the food standards [6]. Keywords—carbon footprint; process optimization; expert The problem of the process optimization is widely known. systems; product life cycle assessment; food processing; global In the agricultural and especially food processing industry warming potential; different techniques are used starting from human-based experience through expert systems to implementing artificial I INTRODUCTION intelligence [6][7][8]. The whole agricultural industry can use United Nations Framework Convention on Climate the whole variety of standards and good-procedures in their Change (UNFCCC) [1], the Kyoto Protocol [2] and the Paris business. The example of such standards might be: Agreement [3] are well known examples that our world and • PAS 2050 [9] - Specification for the assessment of governments are trying to divert climate changes. The climate the life cycle greenhouse gas emissions of goods and changes have taken place several times in the Earth history services ; also in the recent eon e.g. 10000 years in the northern hemisphere. • ISO/TS 14067:2018 [10] -Greenhouse gases - Carbon footprint of products - Requirements and guidelines Nowadays, the climate changes are regarded as one of the for quantification; greatest environmental, social and economic threats facing our planet. It is a result of the industrial revolution and • ISO14040:2006 [11] - Environmental management- statistically shows rapid increase in the average global life cycle assessment: principles and framework; temperature due to the increase in the atmospheric Greenhouse Gas (GHG) concentration, weather changes, • ISO14064-1:2018 [12] - Greenhouse gases - Part 1: draught etc. Specification with guidance at the organization level for quantification and reporting of greenhouse gas emissions and removals. The paper is written as a part of the project CFOOD that is supported by The National Centre for Research and Development, Poland, grant number BIOSTRATEG3/343817/17/NCBR/2018. Copyright © 2019 for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0) In our CFOOD project, the research is aimed at estimating Carbon footprint is typically calculated by considering carbon footprint (CF) for basic basket of frozen vegetable carbon emission factors and activity data, which could be food by applying developed method and software (CF expert evaluated by life cycle assessment (LCA). LCA is based on system, called CFexpert) as well as to develop innovative life cycle inventory (LCI), which is a repository that includes technologies for CF reduction by utilization of vegetable the data of resource and energy consumptions as well as outgrades into valuable products. In the CF calculation task emissions to the environment throughout the global product we take into account PAS 2050 and ISO/TS 14067:2018 to life cycle, see Fig. 1. What is equally important, the problem calculate/estimate CF and later on in the following of uncertainty associated with all phases in the LCI is optimization task of the food processing. important to make LCA-based decisions correctly according to the standards not common sense. For individuals that are curious how to evaluate the CF in their common deeds we can recommend using some formulas In the Tab. I Global Warming Potentials (GWP) [14] used provided by IBM in [13] as well as some CF calculators that for the CF calculations are shown. The values of global can be found in Internet. warming potentials for GHGs to be used in calculations shall be in accordance with Tab. I. The paper is organized as follows. In Section II carbon footprint calculation and its different definition and approaches are presented widely. Life cycle assessment (LCA) and its stages is discussed in Section III. In the next section carbon footprint formulas for the acquisition of raw materials, manufacturing, transportation, usage, and recycle and disposal LCA stages. CFOOD project is shortly presented in the Section 5 as well as the optimization issues emerging and solutions applied to the project. The conclusions are shown in the final section. II CARBON FOOTPRINT CALCULATION In the paper, to estimate carbon footprint (CF) for a given product we take into account PAS 2050 [9] and ISO/TS 14067:2018 [10] as mentioned in Sec. I. . The terms carbon emission and carbon footprint are widely used as an indicator of environmental performance, which is derived from ecological footprint. The carbon footprint of a company, a building, land, a structure, or a retail location is measured in tons/kilograms of CO2 per year, called equivCO2. Product carbon footprint refers to the emission of a variety of GHG gases in a product life cycle. All GHGs specified by IPCC 2007 [14]– includes carbon dioxide (CO2), methane (CH4), nitrous oxide(N2O) plus families of gases like hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), fluorinated ethers (see examples in Tab. I). TABLE I. DIRECT (EXCEPT FOR CH4) GLOBAL WARMING POTENTIALS (GWP) RELATIVE TO CO2 Industrial designation or Chemical GWP for 100-year time common name formula horizon Carbon dioxide CO2 1 Fig. 1. Life cycle assessment (LCA) in carbon footprint (CF) estimation. • Methane CH4 25 • Nitrous N2O 298 oxide III LIFE CYCLE ASSESSMENT IN CARBON FOOTPRINT • CFC-11 CCl3F 4,750 CALCULATION • CFC-12 CCl2F2 10,900 LCA is a widely used approach to assess the actual environmental impact of a product caused by its production • CFC-13 CClF3 14,400 and use. The standards to evaluate the product carbon • Carbon footprint in the LCA are mainly PAS 2050 and ISO/TS CCl4 1400 tetrachloride 14067. The life cycle is defined as a series of consecutive stages where Mi, Gi, Mik, Cik, Gim and GWPim differ at acquisition of of a product by ISO 14040 [12], including acquisition of raw raw materials, manufacturing and transportation stage and materials (in our case vegetable crops), manufacturing (food have different meaning and they are summed up in Tab. I. processing), transportation, usage, and recycle and disposal. Generally speaking: The LCA framework includes the determination of the objective and scope of the evaluation, inventory analysis, life • M stands for materials, manufacturing or cycle impact assessment, and life cycle interpretation [12]. transportation; PAS 2050 uses the LCA framework to evaluate GHG • G is the number of direct GHG emissions at each of emissions from products, either business-to-consumer or these stages and the transportation stage this factor business-to-business. Its main goal is to to minimize carbon as well as the corresponding ones are more footprint. The potential environmental impacts of a sophisticated than in other two stages. production system, either for the entire life cycle of the product or a specific stage, could be effectively assessed In the transportation stage the generated carbon footprint through the LCA of the product. depends on many other factors. The lorries can have different loads, fuels, as well as during the combustion different GHGs In the paper, a carbon footprint calculation is proposed to might be present. It might be summarized by the value of quantify the carbon footprint for all stages of production. activity data at the transportation stage that is estimated for The LCA is divided into four stages, see Fig. 1: i=t as 1. Functional units selection – their selection should be the same for stages of life cycle.  () 2. System boundary determination – to indicate the where: calculation scope; some factors that constitute to less than 1% of total value can be omitted in some cases • Ttj is the quantity of transportation shipment e.g. input of human and animal power. including materials, parts, products, waste, etc. in the k-th transportation stage; 3. Data collection - to calculate carbon footprint include activity data and carbon emission factors in the • Ltk is the transportation distance in the k-th product life cycle as well as their accuracy. transportation; 4. Carbon footprint calculation – it is described in the • EItk is the energy intensity of the k-th subsection II.B. transportation mode. EItk in other words can be briefed as the energy consumption per unit of IV CARBON FOOTPRINT CALCULATION energy quantity and per unit of distance in the k- th transportation mode. According to the definition of product life cycle and the analysis of product carbon footprint given in the PAS 2050 [9], the contribution of carbon footprint is divided into five TABLE II. COEFFICIENTS INTERPRETATION IN ACQUISITION OF RAW MATERIALS, MANUFACTURING AND TRANSPORTATION STAGE stages for the entire product life cycle: acquisition of raw materials, manufacturing, transportation, usage, and recycle Stage Coeffi and disposal. Hence, the total CF for a given product or its cients Acquisition of raw Manufacturing Transportation unit value can be expressed in following formula: materials the number of raw the number of the number of material types manufacturing, transportation r consumed at the processing and stages, including Mi CF =  CF i =a i () acquisition of raw material assembly activity processes road, railway, flight, waterway, etc.; the number of the number of the number of where i is each of the stages of product life cycle, i= a, m, t, u direct GHG direct GHG direct GHG and r are for the acquisition of raw materials, manufacturing, emissions types at emission types at emission types at Gi the acquisition of the manufacturing the transportation transportation, usage, and recycle and disposal stage, raw materials stage and processing stage respectively. stage the consumption of the consumption of the consumption of Carbon footprint of product at the acquisition of raw the k-th raw the energy in the k- the energy in the k- materials, manufacturing and transportation stage can be material th manufacturing, th transportation calculated with very similar formula that is as follows: Mik processing and chain of the assembly activity process processes Mi Gi the carbon the carbon the carbon CFi =  M ik * Cik +  Gim * GWPim () Cik, emission factor of the m-th raw emission factors of the energy emission factor of energy k =1 m =1 material consumed in consumption in the Stage CFi ( p1, p2 ,..., pn ) Coeffi Si = (4) cients Acquisition of raw materials Manufacturing Transportation pi manufacturing, k-th transport mode processing and In the CFOOD project the measure system for the raw assembly process materials and energy resources as well as the transportation the emission of the the emissions of the emission of the are especially prepared. The data from the various elements m-th type GHG at the m-th type GHG m-th type GHG at are united in one information, data acquisition system Gim the acquisition of at the the transportation CFOOD_AS and the knowledge database (KDb). The data raw materials stage manufacturing and stage in the whole processing stage chain about raw materials (vegetables) as well as the usage of some the global the global the global energy resources as coal, gas etc. are inputted by the staff to warming potential warming potential warming potential the KDb system. GWPim of the m-th type of the m-th type of the m-th type GHG GHG GHG in the whole The production line elements are connected to the transport chain CFOOD_AS and the data from the sensors and meters is stored in KDb in the real time. Some data is also derived from the accountant system as shipment data. Carbon footprint of product at the usage and disposal stages can be also calculated in similar way to the previous ones. CONCLUSIONS The CFOOD project is at the initial stage. The whole V. CFOOD PROJECT OPTIMIZATION APPROACH acquisition system is connected and the first real-time tests are conducted. The business partner has started 2019 One of the aims of the CFOOD project is to use outgraded production campaign and the data for products from the materials in the production of the new products e.g. vege- chosen product basket is gathered by the acquisition system burgers. The outgrades can appear at different stages of the and stored in the knowledge database. From the other hand, production line and they are 100% healthy and usable raw the expert system and optimization system is tested and tuned materials that can be used in manufacturing. That is why on two products from the production line. instead of treating them as the waste/disposal they would be used to develop innovative technologies for CF reduction by In 2019 the developed by the authors CFExpert system is utilization of vegetable outgrades into valuable products: planned to be combine with the data acquisition system. The frozen vege-burgers, frozen pastes and lyophilized bars first process optimization will be done to reduce CF in the (lyobars), enriched in fiber, with improved health and products, mainly by lowering the energy resources and water nutritional value. consumption as well as the usage of the outgrades in the new products, that appear during the production in around 5-10%. Different approaches in optimization problem in the measuring CF are used e.g. expert systems, machine learning and artificial intelligence. Well mathematically based ACKNOWLEDGMENT approaches sensitivity analysis is used in [16]. The other TABLE II. We would like to acknowledge the whole approach is green supply chain network design is used [16]. consortium of the CFOOD project especially IBPRS Artificial intelligence and computer vision examples are department from Lodz, that is the leader of the whole project shown in [17]. One of the problems in CF calculations is for their great and successful effort to combine assessment of water usage named water footprint and it is multidisciplinary but very distant research areas in CFOOD shown in [18][19]. LCA approach is also shown in [20][21]. project. Using the sensitivity analysis (SA) in product conceptual design the effect of changing a given input variable or design REFERENCES parameter on a given output of product carbon footprint [1] United Nations Framework Convention on Climate Change. United quantitatively can be measured. Hence, implementing Nations Framework Convention on Climate Change. Retrieved 01 July sensitivity analysis can assess and quantify the uncertainty in 2019. the product carbon footprint. That can also determine the [2] Kyoto Protocol to the United Nations Framework Convention on Climate Change. UN Treaty Database. Retrieved 27 June 2019. impacts of design parameters on carbon footprint in a given [3] Paris Agreement. United Nations Treaty Collection. Retrieved 27 June system [15]. 2019. In that way carbon footprint could be reduced effectively [4] H. C. J. Godfray, Food security: The challenge of feeding 9 billion by revising those most influential design parameters, this people. Science 327, pp. 812–818, 2010. could led to optimization of the each stages of the life cycle. [5] P. Meyfroidt, Trade-offs between environment and livelihoods: Bridging the global land use and food security discussions. Glob. Food In our model used in the CFOOD project, the sensitivity Si of Secur. 16, pp. 9-16, 2018. carbon footprint function described also in the formulas (1)- [6] S. A. Ali, L. Tedone, G. De Mastro, Optimization of the environmental (3) CFi(p1 , p2 , … , pn) with respect to the i-th low-carbon performance of rainfed durum wheat by adjusting the management design parameter pi is calculated according to the formula practices. J. Clean. Prod. 87, pp. 105–118, 2015. [7] M. Kulak, T. Nemecek, E. Frossard, G., Gaillard, Eco-efficiency improvement by using integrative design and life cycle assessment. The case study of alternative bread supply chains in France. J. Clean. Prod. 112, pp. 2452–2461, 2016. [8] M. A. Renouf, C. Renaud-Gentie, A. Perrin, C. Kanyarushoki, F. Jourjon, Effectiveness criteria for customised agricultural life cycle assessment tools. J. Clean. Prod. 179, 246–254, 2018. [9] PAS 2050 (2011) “The Guide to PAS2050-2011, Specification for the assessment of the life cycle greenhouse gas emissions of goods and services. British Standards Institution. [10] ISO/TS 14067 (2018) Greenhouse gases - Carbon footprint of products - Requirements and guidelines for quantification. International Organization for Standardization, Geneva. [11] ISO14040 (2006) Environmental management-life cycle assessment:principles and framework. International Organization for Standardization, Geneva. [12] ISO14064-1 (2018) Greenhouse gases - Part 1: Specification with guidance at the organization level for quantification and reporting of greenhouse gas emissions and removals. International Organization for Standardization, Geneva. [13] Carbon footprint calculation formulas IBM Knowledge Center - https://www.ibm.com/support/knowledgecenter/en/SSFCZ3_10.5.1/co m.ibm.tri.doc/tre_measure/r_carbon_fp_calc.html. Retrieved 27 June 2019. [14] IPCC Guidelines for National Greenhouse Gas Inventories (2006), URL: http://www.ipcc-nggip.iges.or.jp/public/2006gl/index.html, Retrieved 27 June 2019. [15] B. He, et al., Low-carbon conceptual design based on product life cycle assessment. Int. J. Adv. Manuf. Technol. 81, pp. 863-874, 2015. [16] S. Elhedhli, R. Merrick, “Green supply chain network design to reduce carbon emissions.” Transp Res Part D 17, 2012, pp. 370–379. [17] I. Patricio, R. Rieder, Computer vision and artificial intelligence in precision agriculture for grain crops: A systematic review. Computers and Electronics in Agriculture, 153 , pp. 69-81, 2018. [18] Borsato, et al., Comparison of Water-focused Life Cycle Assessment and Water Footprint Assessment: The case of an Italian wine. Science of The Total Environment 666, pp. 1220-1231, 20 May 2019. [19] E., Borsato, et al., Evaluation of the grey water footprint comparing the indirect effects of different agricultural practices. Sustainability (Switzerland) 10(11), Article number 3992, Nov 2018. [20] D. Perez-Neira, A. Grollmus-Venegas, Life-cycle energy assessment and carbon footprint of peri-urban horticulture. A comparative case study of local food systems in Spain. Landscape and Urban Planning 172, pp. 60-68, April 2018. [21] Nabavi-Pelesaraei, S. Rafiee, S.S. Mohtasebi, H. Hosseinzadeh- Bandbafha, K. Chau, Energy consumption enhancement and environmental life cycle assessment in paddy production using optimization techniques. J. Clean. Prod., 162 (2017), pp. 571-586.