Coordinated Integration of Agricultural and Industrial Processes: a Case Study of Sugarcane-Derived Production

The coordinated integration of agricultural and industrial processes in plant-derived production can offer a solution toward sustainability. However, it is hard for general practitioners to realize the coordinated integration of these processes just based on the precedent fact. A special form clarifying the functions of the required activities should be shared among the practitioners for deliberate system design. In this study, a function model for coordinated integration was developed using the type-zero language of integrated definition for object-oriented design. Inputs, outputs, controls, and mechanisms for the required activities and the relationship between them were analyzed through modeling, after which the model was verified based on actual historical facts in the Japanese cane sugar industry. Finally, as a case study from a different industry, the applicability and limitation of the function model in the palm oil industry are discussed. Although the validity of the model should be confirmed through accumulating future case studies, the structure of the function model should be common to industrial crop-derived productions.


Introduction
Sustainable systems design has become increasingly important for the further development of modern society.To achieve sustainable development goals (SDGs) in the next decade, a variety of targets must be addressed, including economic, environmental, and social issues (United Nations General Assembly Resolution 2015).A novel breakaway idea on design thinking could be an important driver for sustainable production (Redante et al. 2019) and should contribute to attaining these multiple goals.The field of process systems engineering has proposed a variety of methodologies and tools for generating novel process systems through design of different objects in chemical engineering and other fields.Warren et al. (2016) developed a product and process design principle, which were designed interactively and efficiently.Stankiewicz and Moulijn (2000) proposed the concept of process intensification, in which the process, equipment, and materials are designed in an integrated manner with the aim of reducing plant size and boosting efficiency.Garcia and You (2016) proposed multiscale modeling ranging from household to global levels to optimize the water-energy-food nexus.It is always necessary to have new perspectives toward a system's boundaries to expand the possibility of finding innovative solutions.
One of the important topics for SDGs is the utilization of plant-derived resources from agriculture, while also leaving room for transformation by expanding the system's design boundaries.Plant-derived production can be divided into agricultural processes at the stage of procuring raw materials and industrial processes at the stage of converting them into final products.Both processes are designed individually.Raw materials tend to be regarded as given constraints in the industrial field, whereas crop production is often discussed within the agricultural field irrespective of how the crop and biomass are processed.Nevertheless, the amount, quality, and seasonality of the biomass, which depend on agricultural factors, affect the performance of the industrial processes.For proper coordination, plant-derived production should be designed considering the agricultural and industrial processes simultaneously.Although various optimization methods have been applied to the biomass supply chain (Sun and Fan 2020), agricultural options such as farming operations have been rarely considered as design parameters.
The potential stakeholders in the coordinated design of agricultural and industrial processes may include researchers in the field, as well as participants in the supply chain: thus, a technology created by researchers based on system-wide thinking would be the key to coordination.The agricultural and industrial sciences are fundamental academic fields that promote agriculture and industry, respectively.Agricultural research has focused on the development of new cultivars and cultivation methods to enhance the productivity and crop quality.By contrast, in industrial science, when the researchers engage in the study of converting biomass resources into a variety of chemicals, they work on the premise that biomass is a uniform material (Isikgor and Becer 2015), although the quality of feedstock sometimes affects its efficiency.Once agricultural and industrial processes are regarded as a unified design target, obstacles that were considered by engineers as constraints might be removed by appropriate exploration of agricultural options, and vice versa.The barriers of emerging technologies to unexplored systems would be lowered as a result of expanding the available technology options.To promote this, collaboration between technology developers and practitioners across the academic boundaries is required.
The cane sugar industry is an example of a coordinated design involving researchers from the fields of agricultural and industrial sciences.Sugarcane has been regarded as a raw material for sugar, but there is an increasing awareness that coproducts from the sugar mills (e.g., bagasse, molasses, filter cake) can be used for a variety of products, such as ethanol, electricity, fertilizers (Gheewala et al. 2011), and chemicals including polymers (Kikuchi et al. 2013;Tsiropoulos et al. 2015).Because sugarcane is related to food, energy, and material production, as well as environmental issues (Silalertruksa et al. 2017), the design of the cane sugar industry attracts much attention as a promising contributor to SDGs.Ohara et al. (2009) reported that in the case of the Japanese industry, the changes in the sugarcane cultivar on the agricultural side and in the millingtechnology option on the industrial side can increase ethanol production while simultaneously maintaining raw sugar production and reducing the environmental burden.Focusing on the cultivar choice as an available option, Ohara et al. (2012) developed a new fermentation technology for cane sugar mills to exploit impurity-rich feedstock for increasing and stabilizing sugar and ethanol production.Ouchida et al. (2017a) argued that integrated modeling of agricultural and industrial processes could aid in finding the Pareto-optimal alternatives based on a holistic perspective.Ouchida et al. (2019) also demonstrated that such a model can also guide the next technology development based on a system-wide perspective.The function of the model was verified as a practical tool for supporting planning for future scenarios by actual decision-makers (Ouchida et al. 2018b).
The agricultural and industrial processes should be designed in coordination for all plant-derived production from agriculture.The above example is a simple historical fact from the cane sugar industry.Other practitioners would find it hard to implement the same practice with only this fact as a reference.The function of required activities and the flow of information among them should be precisely analyzed and shared as a framework.The previous frameworks developed for the sustainable chemical process design (e.g., Liew et al. 2016) cannot sufficiently address the mediation among stakeholders.Several frameworks have been developed to deal with issues associated with multiple stakeholders in the product supply chain.Nakano and Hirao (2008) proposed a framework for the environmentally conscious process design by collaborating with the entire supply chain from the perspective of small and medium-sized enterprises.Uehara et al. (2014) developed a framework for managing environmental and human health risks in middle-stream industries.However, no framework considers the options of both agricultural and industrial processes as design parameters simultaneously considering the differences in their characteristics.
In this paper, a function model of coordinated integration of agricultural and industrial processes in plant-derived production was developed as a framework for design and optimization of their process systems toward sustainability.This framework is proposed as the basis for optimizing process systems between the different industries and to visualize the structure of information exchange and work in optimization.Functions required for general plant-derived production were analyzed based on the concept of the coordinated integration.The type-zero language of integrated definition for object-oriented design (IDEF0) (Ross 1977) was used to model the integrated design activities.The model was verified using actual case studies from the cane sugar industry on Tanegashima, which is one of Japan's remote islands.Finally, the applicability and limitation of the function model are discussed for its generalization.

Function Modeling of Coordinated Integration of Agricultural and Industrial Processes
For an integrated design of agricultural and industrial processes applicable to any plant-derived production, the required activities should be defined explicitly.A previous study showed that an integrated design could be activated by newly developed simulators (Ouchida et al. 2019), technologies, and appropriate experts.However, practitioners in other fields cannot immediately use the knowledge from these cases because the logics that underpin them must be reconstructed and followed.The function of each activity should be generalized to be practicable in other production systems.
One of the most widely used visualization methods for systems design is flow chart depiction (Aguilar-Savén 2004).Although this method visualizes the activities in chronological order, it has a disadvantage in representing the role of information used in each activity.It is desirable for practitioners to understand the desired output to be produced and what information should be used concurrently for every activity.
An appropriate language for depicting the function of activities and flows of information is IDEF0 (Ross 1977), which has been widely used in business process reengineering (Killich et al. 1999), software engineering (Kim et al. 2003;Hwang et al. 2007), and chemical engineering (Theissen et al. 2008).In process design, this method has been used to integrate new or existing engineering methods and tools for environment protection (Fuchino and Shimada 2003) and for evaluation of environmental, health, and safety risks (Sugiyama et al. 2008).
IDEF0 represents activities and related elements of the target system using boxes and arrows (Fig. 1a).The arrow entering the left side represents inputs: that is, items and information converted by the activity.The arrow from the right side represents outputs: that is, products of the activity.The arrow entering the top of the box represents controls: that is, constraints that must be considered for the appropriate outputs.The arrow entering the bottom of the box represents mechanisms: that is, information and resources for the activity as necessary.A model depicted with IDEF0 has a single top activity named A0.The top activity can be decomposed into subactivities, the boxes of which are arranged obliquely and named in numerical order from the top left (Fig. 1b).This layer can be further decomposed into the next level, as necessary.The viewpoint and purpose of the model must be fixed in one model.Actors in the viewpoint conduct the activities.All activities in the model share the same viewpoint.
In this study, we made use of the benefits of function modeling with IDEF0 by considering the following three aspects.First, IDEF0 can be used for generalization by abstracting the relationships between the required activities and information.By bundling information related to an activity into entities, case-specific information can be represented as the difference in instance of the entities, which maintain the generality of the function model.For example, Sugiyama et al. (2008) developed an activity model for chemical process design.While the applicability of this model was confirmed in the case study on methyl methacrylate processes, all chemical processes are within the scope of application.Controls and mechanisms are distinguished in respect of whether the entity is mandatory or not.While practitioners must obey controls, mechanisms are available as needed to promote the activity.When the activity fails, practitioners can negotiate with their source activities to mitigate the controls or analyze the mechanisms that should be procured.In addition, a model can be decomposed into appropriate layers in accordance with an objective.For example, Kikuchi and Hirao (2009) developed a hierarchical activity model for risk-based decision-making processes using IDEF0.The model has five main activities, but the "evaluate process in use" and "decide alternative process" activities are further decomposed because they are the most critical.The interactions between the agricultural and industrial processes should be analyzed in detail.
The structure of the model must reflect the general characteristics of plant-derived production and its validity must be confirmed by actual case studies.To validate the model, the cane sugar industry in Tanegashima was chosen as the case study.Several efforts for coordinated integration of agricultural and industrial processes, such as development of technologies and tools, were performed there.By analyzing these efforts, the kinds of required activities, flows of information between them, and the constraints and available resources in each activity can be validated.Finally, the functionality of activities can be modeled in a shareable form.The developed function model will reinforce the necessities of each activity.By referring to the function model, practitioners can expect to design the systems concurrently considering both the agricultural and industrial options.

Background of Tanegashima
Tanegashima is one of the remote southwestern islands of Japan.Because of its subtropical climate, the primary industry including agriculture, forestry, and fishing have been actively promoted on the island.The primary industry contributed on average 1.1% of the gross domestic production in Japan (Cabinet Office, Japan 2021), whereas it accounted for 9.3% in Tanegashima (Kagoshima Prefecture 2022; Nakatane Town 2020; Nishinoomote City 2021; Minamitane Town 2020).As one of the major crops on this island, sugarcane is converted into raw sugar at one sugar mill and then exported off the island.
The original life cycle boundary of sugarcane-derived production in Tanegashima is shown in Fig. 2.This industry includes sugarcane farming and sugar milling, and has been designed for maximizing raw sugar production.The byproducts of sugar production can be a raw material for ethanol, electricity, fertilizer, and chemicals, but the only product  produced on this island is raw sugar.To protect domestic industries in Japan, sugar production is promoted by subsidies given to sugarcane farmers and sugar mill companies in proportion to the sucrose content.For this reason, the agricultural research institute has developed sucrose-rich cultivars such as NiF8 and NiTn18 ("KF92-93") (Terajima et al. 2010).To maximize sugar yield, the sugar mill adopted the three-boiling system (Yamane 1967), which repeats sugar crystallization three times via recycling seeds and molasses.Sugarcane is harvested from December to April according to its ripening process, which limits the working period of the sugar mill to the same season.The facilities are maintained during the nonharvest season.The timing and the order of farmland for harvesting are regulated by the sugar mill considering the maximization of the yearly sugar production on the island and avoiding inequality among farmers.Sugarcane byproducts have been used effectively.Bagasse is a fibrous residue from sugarcane after sugar extraction.Around 80% of bagasse is consumed inside the mill as fuels.The sugar mill owns a power plant that produces the steam and electricity required for sugar production utilizing the combustion energy from bagasse.This free biomass energy has resulted in rough heat integration and waste of much of the low-grade heat energy under 100 °C (Kikuchi et al. 2016).The surplus bagasse and sugarcane trash are used as bedding for livestock.Filter cake, which is a mineral-rich separation from cane juice, and bagasse ash are returned to farmland as soil conditioners.The final molasses product is exported for other purposes.
Other than the cane sugar industry, there are various other primary and secondary industries.Pasture, sweet potato, and tea have also been cultivated.Parts of the pasture and sweet potato have been in crop rotation with sugarcane to maintain soil fertility.Pasture is fed to the livestock, including milk and beef cattle on the island.Sweet potato is consumed for raw consumption as well as for secondary-processed products, such as starch and shochu, which is a Japanese alcoholic beverage distilled from various ingredients.The factories producing these products are also located on the island and acquire heat sources from fossil fuels.Lumber mills have also been developed using the abundant available forest resources on the island (Kikuchi et al. 2016).
The power system on Tanegashima is an isolated microgrid, which places the island at a disadvantage in terms of thermal power generation, cost, and the security of fossil resources.The cost of energy production is heavy because fuel must be imported from outside the island.The price of light oil in 2019 in Tanegashima was higher by about ¥25/L than the Japanese average (Agency for Natural Resources and Energy, Japan 2021; Kagoshima Prefecture 2021).Although renewable energy has been gradually introduced (particularly since feed-in tariff remuneration was launched for solar energy in 2012), suppression control of such energy has been conducted to avoid power outages (Kyushu Electric Power Co., Inc. 2014).
Regarding the social aspect, Tanegashima faces a shrinking population and aging.In 1980, there were 44,154 residents on Tanegashima, but in 2015, the population had fallen below 30,000 (Portal Site of Official Statistics of Japan 2019).Considering the age pyramid, people over 65 accounted for 34% of the total population in 2015 (Portal Site of Official Statistics of Japan 2019).This situation produced a shortage of famers and successors.

Coordination Tools
There are three tools available to the cane sugar industry in Tanegashima for coordinated integration of the agricultural and industrial processes.The first tool is cultivar development, which is one of the technologies that give birth to cultivars with excellent traits through crossing two different ones.In Tanegashima, the ideal trait of sugarcane is based on the desires of the farmers and the sugar mill.Regarding the ideal traits of sugarcane, farmers desire a high sucrose content, with a high tolerance to insect damage, diseases, and disasters such as drought and typhoons, and harvestability, whereas the sugar mill places emphasis on low content of reducing sugars (i.e., glucose and fructose), minerals, and fibers, which may degrade the efficiency of sugar extraction.Focusing on the function of cultivar development could play a role in the material design for the process systems.
The second is a simulator.Ouchida et al. (2017a) developed a process model that can simulate mass and energy flows in the sugar mill considering both agricultural and industrial options.This simulator is based on a combination of the physicochemical and statistical models.Before the development of the simulator, the relationship between the agricultural and industrial options had been only recognized qualitatively.For example, the flowrate of water for maceration and high-pressure steam for sugar crystallization were adjusted considering the quality of feedstock, which has been intuitively estimated based on the year's climate or harvesting time.The simulator can facilitate the discussion involving both the agricultural and industrial sides through quantifying the relationship.
The third is an evaluation method.For example, a life cycle assessment (LCA) can evaluate the environmental impacts of sugarcane-derived production through its life cycle.An input-output table can estimate the economic impacts of system alternatives on the local society (e.g., Kikuchi et al. 2020;Oshita et al. 2019).Using the aggregated index, practitioners can recognize the impact of their decisions on the system's performance.Although the actual data are not necessarily available for evaluation of the system alternative, the abovementioned simulator can be used to provide data for the rough estimation.

Function Modeling Result by IDEF0
Analyzing the historical facts, a function model for the integrated design of agriculture and industry was developed using the IDEF0 modeling language (see Sect.Function Modeling of Coordinated Integration of Agricultural and Industrial Processes).The top activity (A0: Design agricultural and industrial processes in an integrated manner) is shown in Fig. 3.The viewpoint is defined as a consortium of producers in plant-derived production.The principal outputs of this activity are practical system alternatives for the design of agricultural and industrial processes, accompanied by the technologies required for achieving them.The diagram identifies the players required for implementing these alternatives.If the consortium fails to generate the above output, they request that the other stakeholders reconsider the constraints toward the consortium, so that a middle ground between them can be reached.
The activity controls are classified into five categories shown in Table 1.The consortium is tasked with transforming "Present systems" into improved ones."General constraints" such as geographical features and climate conditions are uncontrollable, irrespective of the consortium's actions."Constraints from externals" are determined by the higher decision-makers or by external ones.For example, the budget available to the consortium or product functions are limited by the company's decisions or by existing customers' demands, which could be modified by negotiation."Constraints from internals" are judgment criteria or values of the consortium contributors.This project could originate from either of two types of constraints: "projectspecific constraints" or "evident problems."If the project is  launched as a routine work, "project-specific constraints" are given and followed.Otherwise, the project starts to cope with "evident problems." The activity is performed with the support of some mechanisms.Stakeholders are involved in this activity, as necessary.The consortium consults appropriate experts if some technical knowledge is needed.The mechanisms required for achieving this activity are developed with the help of mechanism developers.Existing information resources, evaluation methods, technologies, and models are consulted in aid of all the activities.
The subactivities for activity A0 are shown in Fig. 4. Activity A1 ("Manage the design of agriculture and industry") is performed to control all other activities and output final products.It interprets control arrows from outside the project, which might be vague and undefined in the early stage, to convey clear information to other activities, such as project objectives, spatiotemporal boundaries, information about the present system, and definitions of required mechanisms.Activity A1 also receives feedback from other activities within the project as control arrows.Finally, the products of the project are outputted after collecting information.Based on the controls provided by activity A1, the present system is evaluated in activity A2 ("Evaluate the present system").Activity A3 ("Generate alternative candidates") generates alternative candidates of the system for comparison with the present system.This prominent activity enables simultaneous consideration of both agricultural and industrial options to search a broader space of alternatives.Activity A4 ("Evaluate alternative candidates") evaluates the alternative candidates generated in activity A3 according to the way the present system is evaluated.Although activity A4 has much in common with activity A2 regarding evaluation procedures, they differ in their data acquisition methods.To estimate the process inventory, a technological system simulator is required as an additional mechanism.Activity A5 ("Decide on alternative systems") screens alternative system candidates comparing the present system with alternative candidates.The required mechanisms for the above activities are provided from activity A6 ("Prepare mechanisms").Activities A1 and A6 are different from other activity models.Activities A2, A3, and A4 were discussed in the framework of systems design (Sugiyama et al. 2008;Kikuchi and Hirao 2009); therefore, their detailed analysis is of little value.Decomposing the A5 layer is also of little value because it is a matter-of-fact procedure.By contrast, activities A1 and A6 should be analyzed in detail.A1 is a management activity that should integrate agricultural and industrial processes, whereas A6 is a research and development activity that organically prepares three different mechanisms.The model in Fig. 4 is insufficient for representing the intricate communication between A1, A6, and the other activities; therefore, they must be further decomposed into their subactivities.
Activity A1 ("Manage design of agricultural and industrial processes") is decomposed in Fig. 5. Activity A11 ("Set project objectives") converts constraints from the upper layers into clear project objectives, such as assessment index and the desired standard of system performance.Based on the objectives, the spatiotemporal boundaries are determined in activity A12 ("Define system boundary").Activity A13 ("Scrutinize present system") checks the information about the present system such as mass and energy flow data in the defined boundary.A14 ("Define required mechanisms") is the key activity that defines required mechanisms based on the request for developing new mechanisms and generated alternative system candidates.In the conventional way, agricultural and industrial options are regarded as constraints for the industrial and agricultural sides, respectively.However, A14 can consider both options as design parameters breaking their inter-constraints from the other side, and expanding the possibility of outstanding output.Screened alternative candidates are obtained through the interactions between activities A11-A14 and outside of A1.Finally, the practicability of the candidates is judged to output practical system alternatives in activity A15 ("Decide on alternative systems").The decomposition of activity A6 ("Prepare mechanisms") is shown in Fig. 6.Activity A61 ("Specify required mechanisms") formalizes the requests for developing mechanisms considering existing technologies, models, and evaluation methods that meet their definitions.If there are insufficient mechanisms, the requests for developing new mechanisms are outputted.Activities A62-A65 develop new mechanisms including technologies, models, and evaluation methods under the control of the requests.It is critical to develop agricultural technologies (A62) and industrial technologies (A63) in parallel.With the coordination of agricultural and industrial requirements and constraints in A14, these activities have the potential of creating outstanding new technologies.These mechanisms are referred to again in A61.The specifications of required mechanisms are conveyed to A66, where the required mechanisms, including the existing and the one developed, are provided outside of A6.If activity A61 fails, the failure is outputted.

Historical Facts for the Coordinated Integration of Agricultural and Industrial Processes on Tanegashima
The applicability of the function model was verified using an actual case study.The history of the cane sugar industry in Tanegashima was chosen as a case for this study.In this case, the viewpoint was set as the Shinko Sugar Mill Company (Shinko), which is the sole manufacturer of raw sugar from sugarcane in Tanegashima.
Research and development for biomass utilization by avoiding fossil resources has been progressing rapidly and cane sugar mills have been regarded as holders of large biomass, such as molasses and bagasse.Therefore, they are a good target for applying the new technologies.Since the 2000s, under the control of profit improvement (C1) and social needs (C3), Shinko has decided to collaborate with Asahi Group Holdings Co. Ltd. (Asahi), which is a manufacturer of beer and beverages, to pursue the effective use Fig. 6 A6 subactivities (modeling result) of molasses (A1).Asahi is an expert in fermentation technology (M3) and also a fermentation technology developer (M7-1).Before the collaboration, Shinko regarded molasses as a waste product.Asahi's fermentation technology had the potential of producing ethanol from molasses instead of paying disposal costs.After the effective use of molasses was set as an objective (A11), the molasses-fermentation technology was defined as a required mechanism (A14) and Shinko started to develop the process with Asahi (A6).
There were still two challenges for installing the fermentation technology in actual cane sugar mills (Ohara 2013) .The first was social acceptance (C3).In general, food is given preference over other products.In the case of the cane sugar industries, bioethanol should be produced without disturbing sugar production.In fact, the subsidies for sucrose-rich sugarcane have become a barrier to the coproduction of other products in Japan.The second was the economic stability of the cane sugar industry (C4).The economic conditions fluctuated easily because of disasters and the market price of sugar and oils.There is a high hurdle for industries to make an additional investment in unstable productions.These challenges were recognized as definitions of required mechanisms (A4), which the fermentation technology alone cannot live up to (A61).
To overcome these challenges, ethanol production was added to the project objectives (A11), the system boundary was expanded to farming (A12), additional data and information were collected through scrutinizing the systems (A13), and the selection of high-yielding sugarcane was added as one of the available options based on the existing reports (M1).Applying the same amount of fertilizer and the same environment, high-yielding sugarcane generally accumulates a larger amount of sucrose per unit area than other cultivars.It also has a relatively high tolerance against typhoon-caused stalk breakage.Despite these good traits, the use of high-yielding sugarcane in sugar mills was unpopular because it tends to store reducing sugars, minerals, and fiber, which may degrade the sugar yield.To analyze the performance of using high-yielding sugarcane quantitatively, Ohara et al. (2009) conducted an LCA using the process inventory obtained from the actual sugar mill (A2).The result showed that ethanol production from highyielding sugarcane can reduce the life cycle greenhouse gas (GHG) emissions without degrading raw sugar production.This result inferred a leap in the system's performance by retrofitting the design for multiple production.Modifying the objective of cultivar development by defining sugarcane not only as the raw material for raw sugar but also for ethanol, the high-yielding cultivar "KY01-2044" was developed (A61, A62, A66) through collaboration with the National Agriculture and Food Research Organization (NARO) (Ouchida et al. 2019) (M7-1).A pilot-scale experiment using KY01-2044 indicated an increase in both raw sugar and ethanol (Ohara et al. 2018).
Although KY01-2044 has become an alternative cultivar candidate (A3), there was still the challenge of its higher reducing sugars content.Once reducing sugars degrade the sucrose yield in the sugar crystallization process, a delay in batch scheduling may follow and, consequently, a drop in throughput (Ouchida et al. 2017b).Because there was no solution mechanism for this issue, a request for developing new mechanisms was returned to A1. Sharing the new requirement with fermentation experts, the concept of an "inversion process" (Ohara et al. 2012) was created in A61.In this concept, only reducing sugars are selectively converted into ethanol, which is removed with water in the following evaporation process, and finally sugar is crystallized efficiently in the sucroserich solution.Based on the specified required mechanisms, an inversion process was developed (A63, A66).
Through the development of KY01-2044 and the inversion process, the requirement for process systems was reviewed: i.e., agricultural and industrial processes should be designed in an integrated manner.In fact, various options in farming stages such as the choice of cultivar and farm operations affect the yield and composition of the harvested sugarcane, which may in turn affect the processes in industrial stages.Involving researchers and engineers as mechanisms of model developers, a systems model that can simulate mass and energy flow considering both agricultural and industrial options was developed (Ouchida et al. 2017a) (A61, A64).This systems model was extended to consider the inversion process (Ouchida et al. 2019) and the replacement of the steam turbine in the power plant (Ouchida et al. 2018a).
The systems model can be an important tool as a technological system simulator for supporting the research and development for both the agricultural and industrial aspects through quantification of some indexes (A4).In fact, a new cultivar called "Harunoogi" was developed by NARO in 2019 (Hattori et al. 2019).The four major sugarcane cultivars available in Tanegashima are summarized in Table 2.The production balance of raw sugar, ethanol, and excess electricity for the four cultivars as shown in Fig. 7.This graph indicates that a cultivar selection based on combined production can increase raw sugar production by 1.48 times at maximum from single raw sugar production case using NiF8.Based on the LCA, besides productivity, Harunoogi has secondary advantages in profitability for the farmers, the sugar mill, and GHG emissions (Figs.8-10 in the Appendix, respectively).Comparing single production targeting raw sugar using NiF8, combined production targeting the three products using Harunoogi pulls up annual profits for the farmers by 1.934 billion JPY and for the sugar mill by 869 million JPY.Moreover, 0.746 kg-CO 2 eq.kg-sugar −1 can also be saved mainly owing to the avoidance of fossil-based electricity production by using surplus bagasse as fuel.
The technological system simulator enabled a new collaboration between the cane sugar industry and the surrounding area when generating alternative candidates (A3).One of the examples is the idea of interconnection of the power plant in the sugar mill with the island power system.The grids of the southwestern Japanese islands, including Tanegashima, were subsidized because of their disadvantage in purchasing fossil fuels and their indispensability.After 2016, a full liberalization of the electricity retail market was started, and the traditional grid company was expected to lose the incentive of generating electricity with low returns in the remote islands.According to Kikuchi et al. (2016), however, there could be another power source that utilizes abundant biomass resources such as wood chip.If the biomass resources are fed to the power plant in the cane sugar mill, more electricity would be generated, which could be more flexibly produced if a high-yielding cultivar were chosen and back-pressure steam turbines replaced with condensingextraction steam turbines (Ouchida et al. 2018a).As a technology to close the temporal and spatial gaps between the suppliers and consumers, a zeolite boiler, which can reduce the consumption of fuel by the thermal energy storage system utilizing the water vapor ad/desorption cycle of zeolite (Fujii et al. 2019) and life cycle GHG emission (Fujii et al. 2022), was developed (Fujii et al. 2016) (A61, A63, A66).The economic effect of such technology options on the local economy was analyzed (Kikuchi et al. 2020;Oshita et al. 2019) (A61, A65, A66).Finally, a meeting was held to discuss alternative systems considering both agricultural and industrial technology options using the simulator (Ouchida et al. 2018b) (A4, A5, A15).

Outcome of Function Modeling
The function model gives designers in plant-derived production three indications.The first is the effectiveness of expanding the systems boundary.In general, when we tackle problems we face, we tend to seek the solution within the systems considering the options that we own.However, other outside solutions may emerge through collaboration with others.Although the importance of external communication among stakeholders for sustainability has been acknowledged (Azapagic 2003), few case studies have been reported to confirm this recognition.In cases of the relationship between agriculture and industry, the approach of seeking solutions in collaboration with external bodies could be affirmed.
The second is the value of modeling agricultural processes with industrial ones.The modeling approach for simulation is known to be a promising methodology for decision-making in sustainable systems design (Gbededo and Liyanage 2020).The industrial process has been a major target for a simulation-based approach, and for which the framework of simulation-based systems design was developed (Sugiyama et al. 2006(Sugiyama et al. , 2008)).The agricultural process, on the other hand, has not been a common target of simulation-based systems design because of the uncertainty of environmental conditions and the temporal scale of  farming.Eventually, heuristics have often been superior to the simulation.However, the function model makes us realize the new potential of the system simulator as a systems design tool for expanding our vision.Despite the physical and spatiotemporal differences between the agricultural and industrial processes, both can be designed with the provision of appropriate mechanisms and through repetitive activities.
The third is the functionality of the abovementioned simulator as a discussion tool among multiple stakeholders.Information sharing among supply chains is crucial to their sustainability and, hence, software development to enable such task has drawn attention (Papetti et al. 2019).Systems designers sometimes vaguely intend to collaborate with other stakeholders over the business boundary, but do not have an approach to perform the next action because of the absence of common languages.In the case of plant-derived production, the difference in academic fields might have been a major hurdle for communicating with each other.The simulator can function as a mechanism for promoting the discussion over business boundaries.The function model shows that the systems design incorporating systems modeling could create an outstanding system alternative and technology through reforming the consciousnesses of multiple stakeholders.
As a result of systems design following the function model, a Pareto-optimal solution might greatly move forward the design, driven not only by the inner system but also by an outer system.Once the relationship between agricultural and industrial processes is restructured based on the proposed concept, the viewpoint of stakeholders toward the systems could change, resulting in the modification of the control of the project.For example, in the case of the Japanese cane sugar industries, the rule of subsidies for sugarcane farmers may change.Currently, to motivate sugar production, the subsidy increases in proportion to the sucrose content of sugarcane, although this rule paradoxically hinders the increased production of raw sugar through the integrated design.One of the reasons for the opposite perspectives is the miscommunication among stakeholders.The technological system simulator expects to lower the hurdle for the Japanese government to modify the rule by showing the quantitative ground that a sugarcane cultivar with low sucrose content, but high yield can increase and stabilize raw sugar production.Namely, the function model represents the possibility of enhancing Pareto-optimal solutions through modifying the controls (C3 in Fig. 3).

Distinct Feature of Integrated Design of Agriculture and Industry
The frameworks for some systems designs have been developed previously.Sugiyama et al. (2008) developed a design framework of a chemical process to consider the environment, health, and safety.The designers should pay attention to the following points: generation, evaluation, and selection of alternatives are performed with the aid of providing mechanisms, such as process model and evaluation methods; estimated mass and energy balances in a static state are available for the evaluation in the early design stage; the data and methods used should be reviewed and updated in accordance with the design stages.The same logic can be partly shared with the integrated design of agricultural and industrial processes.However, it should have new considerations in two aspects that are far different from the chemical process design.
The first is the subject of a project.In contrast to chemical process design, the subject of which is fixed to a project manager in a single profit-seeking organization, a project member of the integrated design should be a consortium (e.g., it can invite stakeholders in all related fields as new members and modify their views flexibly).Although the whole supply chain of a plant-derived production would be manageable by the most downstream supplier, a project manager in a single profit-seeking organization may not thoroughly consider the interests of all related stakeholders.A consortium should be organized to manage the project reasonably.
The second is the driving force for launching the project.It is not always possible for a project manager to recognize the importance of an integrated design of agricultural and industrial processes before beginning the project.It would be more likely to start a project by examining a countermeasure for actual on-site problems like productivity enhancement and stabilization.It is not until the part of the problems is attributed to other stakeholders' decision-making outside their field that there would be sufficient motive for an integrated design approach.The procedure should be represented so that a project can be initiated by either or both project-specific constraints and apparent problems.
As a next step after the systems design, an additional characteristic task of planning future scenarios would be required.It is hard to immediately shift the present systems to new ones because of the temporal gap between agricultural and industrial options.For example, the constitution of crops in a region cannot be changed until at least the next season.In general, farming is managed in a cycle that varies from a season to several years.This temporal scale largely depends on nature and is hardly controllable, unlike industry.This gap demands the project to set an additional task of planning scenarios to avoid obstacles in shifting the systems.Because the order of options could affect interests among stakeholders, a consensus about their timing should be built before the implementation stage.At this time, stakeholders will become players by receiving the requirements output from the design procedures (O3).Stakeholders who have been in vague and unclear relation with the project may change their awareness as players who make real decisions.Such players, as well as the new technologies, would be a key for the design of the agricultural and industrial processes in the next step.

Applicability and Limitation of the Function Model as a Framework for Coordinated Integration of Agricultural and Industrial Processes
The collaborative approach involving both agricultural and industrial processes is of great value for a variety of plantderived products.Biorefineries that process industrial crops as feedstock strongly desire a novel design methodology that achieves their high-level collaboration.In a comparable way to oil refineries, biorefineries convert biomass as a mixture of various chemical substances into multiple value-added products.However, the systems design concept for oil refineries is insufficient for industrial-crops-derived biorefineries because the agricultural processes are out of the design scope.There is not much worth in designing the material acquisition stage in oil refineries, but it is valuable for biorefineries to design agricultural processes strategically, because in many cases they are originally designed according with other intentions, such as, for example, maximizing the initial product or just the agricultural crops, despite their large impacts on the performance of the whole system.The function model developed in this paper has the potential of facilitating the transition to the holistic systems design involving both agricultural and industrial processes.
The function model was built based on the cane sugar industry; therefore, its applicability to other cases should be discussed.Considering the growing trend of production, in this section the palm oil industry is chosen as an example to discuss the applicability.Oil palm is a tropical agricultural crop grown to satisfy the demand for vegetable oils.The palm oil industry has shown an outstanding growth during the last two decades, particularly in Eastern Asia (Food and Agriculture Organization of the United Nations 2021) and is expected to continue increasing (Sumarga and Hein 2016).It has similar characteristics to sugarcane in the process flows.Farmers harvest both crops before they are processed in a factory adjacent to the farmland.Palm oil production accompanies several byproducts like bagasse or filter cake in cane sugar mills.Empty fruit bunches (EFB) are a fibrous byproduct that is used as a fuel for the internal boiler or mulch.Palm oil mill effluent (POME) is a liquid waste stream from palm oil mills that is treated to generate biogas that can be used as energy resources in some mills (Schmidt and Rosa 2020) .
Oil palm and sugarcane are also similar in their sustainability issues.To meet with the growing food and bioenergy demand worldwide under limited arable area, a solution for improving their productivity is urgently needed.Attention is increasingly paid to the environmental burdens for their cultivation and processing.The deforestation by land-use changes to oil palm plantations have been major contributors to GHG emissions of palm oil production (Lam et al. 2019) and have caused loss of biodiversity.The treatment of byproducts like POME in a palm oil mill and filter cake in a cane sugar mill are studied to reduce the environmental loads.As social issues such as human rights conflicts are disputed, the Roundtable on Sustainable Palm Oil (RSPO) developed and released a voluntary certification schema in 2008 to suppress such sustainability issues in palm oil production (Roundtable on Sustainable Palm Oil 2021).
The properties of oil palm-derived production and the systems design activities can be explained along with the function model.The sustainability issues including the environmental and social problems that emerged in the palm oil industries are placed on C1 ("Evident problems or project-specific constraints").They can be the initial motivations for launching a project to get a solution.As a social issue, the promotion of oil palm plantations sometimes conflicts with human rights in rural areas.Their values are in C2 ("Constraints from internals") that should be free from intrusion in the project.External supports such as subsidies for oil palm productions and RSPO certification are categorized in C3 ("Constraints from externals") that may give the project incentive to solve the problems in C1.C4 ("General constraints") includes climate conditions and crop-specific constraints.
The past research activities in palm oil industries can be positioned on the parts of the function model.The reports about LCA of products from oil palm (e.g., Carlson et al. 2013;Gamaralalage et al. 2022;Lam et al. 2019;Xu et al. 2020) originated from the evident problems that palm oil production may have caused serious environmental issues.The increase in GHG emission that resulted from the change of land use for the expansion of palm oil were visualized as scientifically informed data for Kalimantan (Carlson et al. 2013) and other regions in Indonesia (Lam et al. 2019).The GHG emission from the cradle-to-gate life cycle of palm fatty acid distillate (PFAD) should be carefully examined to consider the classification of PFAD in the life cycle of palm oil, namely, residue, byproducts, or coproducts of palm oil processing (Xu et al. 2020), the environmental loads of which include the processes from palm cultivation to palm oil refinery.The settings in LCA on palm oil plantation has also major impacts on the GHG emission reduction potential of the application of palm kernel shell (PKS) as biofuel in power generation (Gamaralalage et al. 2022).In these studies, project objectives were set considering every kind of constraints (C1-C5) in A11 to define the system boundary (A12) and to scrutinize the present system (A13).Because of the absence of tools for discussing the ideal process systems, the processes in the palm oil mill stage were modeled (A14, A61, A64) (e.g., Schmidt and Rosa 2020) .The provision of the model (A66) enabled the evaluations of the present system (A2), the generation of alternative candidates (A3), and the evaluation of these alternative candidates (A4).
The research activities to date will reach the integrated design of agricultural and industrial processes by tracing the function model.For example, oil palm cultivar options can be discussed with industrial technology options as is the case in sugarcane-derived production.High-yielding oil palm has been developed aiming maximization of palm oil production (e.g., Yarra et al. 2019).A64 will extend the process model that can consider cultivars, including unpopular ones, with industrial technology options by collecting data about their yielding and compositions as existing information.In addition, the requirements for cultivars and industrial technology options can be redefined thanks to the extended model.The cultivar change will lead to the change in the amount of byproducts from oil palm milling, which may have impacts on the mass and energy balance in the mill.For example, lignin can be extracted from EFB, palm mesocarp fiber, and PKS, and its extraction efficiency highly depends on the initial lignin content in oil palm (Rashid et al. 2018).The new characteristics of oil palm may be required if cultivar choices and mass and energy balance to produce oil palm, fuels, and other products in industrial processes are considered simultaneously.Through propelling the project along the activities in the function model, oil palm-derived products can also expect to find wider solutions.
The function model can be applied to productions from industrial crops, such as sugarcane and oil palm, but there are other plant-derived resources that are expected to be applied but require further discussion.There is still a question as to the application to the production from woods because the characteristics of the material acquisition stage in forestry might be different from those in agriculture.Because woods grow across a generation, organizing a consortium among stakeholders for the integration of forestry and industrial processes might require additional activities to overcome the huge temporal gap.In contrast to sugarcane and oil palm, cereal crops such as rice, corn, and barley may also have various aspects to be considered as the raw materials for plant-derived production because the added value of their main products are greater than the obtained biomass from such crops.The most of biomass from the industries related with such cereal crops has been regarded as residues and used as raw materials for chemical, fuel, and energy generation such as rice husk, straw (Lim et al. 2012), and bran oil containing oleochemicals (Zhang et al. 2021), corn stover (Ahorsu et al. 2018), and yeast extract residue (Shimada et al. 2021).Currently, grain farmers have little incentive for the effective use of these biomass residues because the mutual dependency between the agricultural and the industrial processes for biomass utilization is not always strong.Depending on how boundaries are defined, as seen in the PFAD example (Xu et al. 2020), the farmers can be motivated through the effective use of biomass.For this purpose, the proposed function model could be applicable as a framework to facilitate the coordinated integration of agricultural and industrial processes.Actual case studies are required to validate the function model in such a case.

Conclusions
A function model of integrated design of agriculture and industry in plant-derived production was developed for object-oriented design.A set of activities for the coordinated integration was visualized with the flow of information between them.Through function modeling, it became clear that generation, evaluation, and screening of alternative candidates are advanced accordingly with the help of prepared mechanisms including technologies, a technological system simulator, and evaluation methods.There is a seed of sustainable solutions in the cycle of developing agricultural and industrial technologies in parallel and defining required mechanisms, where both options are simultaneously regarded as design parameters.
After developing a function model for general plantderived production based on the proposed concept, the model was verified through actual case studies in the Tanegashima cane sugar industry.In this case study, the Japanese cane sugar industries have created several prominent technologies, such as a high-yielding sugarcane cultivar as an agricultural option and selective fermentation as an industrial option.The creation of such new technologies addresses the value of a quantitative discussion that is enabled by integrated modeling of the agricultural and industrial processes.
The function model was found to be applicable to industrial-crop-derived productions through case studies on the cane sugar and palm industries.In general, industrial crops are cultivated on the premise that they are changed into value-added products via factories; therefore, there should be mutual dependency between the agricultural and industrial processes.Because the structure of the function model would be common among industrial-cropderived productions, it would lower the hurdle for other practitioners to generate sustainable system alternatives.However, the applicability of the function model has not yet been verified for all plant-derived productions on cases like woods and cereal crop-derived productions.Wood-derived production needs careful discussion on the temporal gap between forestry and the industrial processes.In the case of cereal crops, processing biomass residues is not a prerequisite for farmers.There may remain some properties that have not yet been discussed in such cases.Currently, sustainable business models which connect to innovative technologies and social changes are strongly needed, whereas there are few actual case studies (Bradley et al. 2020).Validation of the function model through accumulating case studies would clarify the applicability and bring practical insights for achieving sustainable production in a variety of businesses related to plantderived production.

Appendix
Figure 7 in Sect.3.2 is the simulation result from the systems model developed in other works (Ouchida et al. 2017a(Ouchida et al. , 2019) ) using the data on Harunoogi (Hattori et al. 2019).The data and parameters to output the simulation result can also be referred from the previous works (Ouchida et al. 2017a(Ouchida et al. , 2018a(Ouchida et al. , 2019)).The profit and expense for farmers and the sugar mill, and the cradle-to-gate GHG emissions of raw sugar based on LCA for combined sugar, ethanol, and electricity production are also shown in Figs. 8, 9, and 10, respectively.Regarding Figs. 9 and 10, evaluation results of cases where only sugar is produced as a final product, as was the case in Tanegashima were added for comparison.
All conditions for the evaluations follow the previous work (Ouchida et al. 2017a) with the provision that the area of farmland was 2500 ha, the cropping type is spring-planted in 25% of the area and ratoon in the rest, the addition rate of maceration water in the milling process was 0.3 kg water/kg fiber, and the repetition count for sugar crystallization was three.In the combined production cases, the currently used back-pressure steam turbine in the power plant of the sugar mill was replaced with a condensing-extraction steam turbine (Ouchida et al. 2018a).Throughout the working period of the mill, the surplus electricity after the sugar and ethanol production was sold to the grid.Products and byproducts were purchased in the amounts produced from the harvested sugarcane.The unit price of raw materials and products, GHG emission inventory for each material, and the rule for determining a sugarcane price also refer to the previous work (Ouchida et al. 2017a).The costs for the construction of an ethanol plant and the steam turbine replacement were excluded from consideration to evaluate under static conditions for comparison.

Fig. 2
Fig. 2 Original life cycle boundary of sugarcane-derived production on Tanegashima (created based on interviews to the local practitioners)

Fig. 7
Fig. 7 Production of raw sugar, ethanol, and electricity from four different cultivars, estimated in this study.Size of the circles indicate relative production of electric energy

Table 2
Characteristics of sugarcane cultivars grown on Tanegashima Spring-planted: crops are planted in spring and harvested in the next spring.Ratoon: crops are grown from roots remaining underground after the last harvest.The data presented is the average from 2015 to 2017 for spring-planted and from 2016 to 2018 for ratoon