Enhancing Frozen Food Production Circularity with Systematic Innovation

Abstract

The frozen food industry faces growing pressure for sustainability, requiring significant reductions in energy consumption and environmental impact. A critical, yet often overlooked, challenge is the temperature differential between production equipment and the environment, which causes energy inefficiencies, material losses, and quality degradation. Despite its significant influence on production performance and environmental burden, this issue has received limited attention in existing studies. To address this gap, this study aims to develop a systematic innovation framework that integrates the TRIZ-based system interaction analysis, a knowledge base of patented and expert-derived solutions, and Quality Function Deployment to identify root causes and design effective circular strategies. The proposed framework is applied to a frozen fish processing case study, where analysis reveals temperature variation as the dominant bottleneck. The strategy embeds dynamic, microscale adaptability into freezing equipment, which reduces the environmental impact of temperature fluctuations, improves energy efficiency, and cuts material waste. These results demonstrate the feasibility, scalability, and innovation potential of the approach, offering a structured methodology for advancing circularity, resilience, and eco-efficiency in frozen food processing systems.

1. Introduction

The frozen food sector plays an increasingly important role in the global food industry, offering a cost-effective and nutritionally stable alternative to fresh food products [1]. Its ability to reduce spoilage and extend shelf life has made it a practical solution to food waste reduction [2,3,4]. Driven by growing demand for preservative-free products and long shelf-life options, the market is expected to exceed USD 430 billion by 2030 [5,6,7]. Despite this growth, the industry faces significant challenges: it is highly energy-intensive and environmentally burdensome, contributing substantially to global energy consumption (approximately 8%) and greenhouse gas (GHG) emissions [8,9,10]. The carbon footprint of certain frozen products can be alarmingly high, with temperature control technologies accounting for 30–50% of the total [11]. Moreover, maintaining temperature stability is critical for product quality and safety; even minor deviations can drastically reduce shelf life and increase the risk of spoilage and foodborne illness [12]. The industry is also increasingly vulnerable to both short-term environmental shocks and long-term climate challenges [13]. While prior studies have addressed preservation technologies and food waste valorization [14,15,16,17], a crucial research gap remains: the dynamic interactions between equipment and environmental temperatures have been largely underexplored, despite their significant impact on energy use, food quality, and process reliability.

To address these pressing challenges, the circular economy (CE) has emerged as a key strategy for sustainable food systems, emphasizing resource reuse, waste reduction, and regenerative production cycles [18,19,20]. CE practices aim to enhance eco-efficiency and reduce GHG emissions in food processing by redesigning processes to minimize non-renewable inputs [21]. However, applying CE in complex industrial systems like frozen food production requires a structured and innovative approach. To address these challenges, the Theory of Inventive Problem Solving (TRIZ), developed by Genrich Al’tshuller in 1946, provides a robust framework for this purpose [22]. TRIZ offers a structured methodology to identify and resolve engineering contradictions by analyzing patterns in patented solutions. It has been widely applied across engineering, product design, and management [23,24,25] and increasingly adopted for sustainability innovation [26,27,28]. Key tools such as Function Analysis (FA), Substance-Field Analysis (SFA), and the 76 Standard Solutions identify negative interactions and propose targeted, knowledge-based improvements [29,30]. When combined with Quality Function Deployment (QFD), this methodology ensures that technical solutions align with practical, stakeholder-driven requirements [31].

This study proposes a systematic innovation framework grounded in TRIZ to facilitate the circular transformation of frozen food production. Our framework addresses the aforementioned research gap by analyzing interactions within processing systems using function-oriented modeling aligned with CE principles. In addition, a case study in frozen fish production will be used to validate the framework to demonstrate a reduction in environmental impact caused by ambient temperature variation and measurable gains in energy efficiency and waste minimization. Therefore, the objectives of this study are to: (1) identify the root causes of inefficiency through FA and SFA, (2) filter and select appropriate standard solutions from the 76 Standard Solutions in a TRIZ database, and (3) apply QFD to prioritize solutions based on their alignment with both technical needs and circular economy goals. Overall, this research presents a novel integration of TRIZ, CE indicators, and QFD to address a long-standing but under-examined challenge in the frozen food industry. The resulting framework offers a replicable and scalable approach to designing circular production strategies, fostering innovation in energy-intensive food processing systems.

2. Materials and Methods

This study proposes a structured methodology that integrates systematic innovation with food-specific circular economy principles, facilitating the circular transformation of frozen food production systems. The research framework is organized into two main stages: problem definition and solution generation, as shown in Figure 1. The first stage identifies the root causes of production bottlenecks using CE characteristic indicators and function-oriented structural modeling. The second stage then applies the TRIZ framework, specifically using SFA to model core problems and match them with appropriate inventive strategies from the 76 standard solutions. To ensure practical relevance and prioritization, QFD is used to rank and refine the selected solutions based on CE performance indicators. A detailed discussion of this framework and its stages is provided in the following sub-sections.

2.1. Problem Identification for Circular Transition in Frozen Food Production

This section utilizes CE food characteristic indicators and Function Analysis (FA), a TRIZ tool that illustrates system interactions, to diagnose inefficiencies and bottlenecks in frozen food production. The steps are as follows:

Step 1: Establish the function model based on the food processing operation chart

According to the selected product production process (i.e., frozen fish processing in this paper), this research first constructs a detailed operational flow chart to describe the current production procedures. Then, a function-oriented modeling approach is employed to analyze system interactions within the production line. This function model illustrates the relationships between system components and helps identify the problems arising from the interactions. Each function is then assessed and categorized as insufficient, excessive, or normal based on the Function Analysis (FA) method [32,33,34]. While FA is widely used in product design, its application in production process optimization remains limited. To enhance the applicability of FA on the production aspect, this study redefines the concept of ‘function’ as:

“Function: The specific operation performed by an equipment, operator, or other production elements (the function carrier) on an input material, product, or other objects to achieve a desired effect or change in its parameters, contributing to the overall production goals and efficiency.”

In this model, each function is represented as consisting of the function carrier, the action, and the object, as expressed in Equation (1).

$$\mathrm{F}\mathrm{u}\mathrm{n}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}=\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n}\to \mathrm{o}\mathrm{b}\mathrm{j}\mathrm{e}\mathrm{c}\mathrm{t}$$

(1)

Environmental factors have a more significant impact on the frozen food industry compared to other sectors, particularly regarding temperature sensitivity and the risk of rapid quality degradation due to climate variation. In this context, this study analyzes the production environment through field surveys conducted in a selected processing plant. The super-system elements representing environmental conditions are integrated into the function model to capture their influence.

Through this step, the study constructs a comprehensive representation of the production system, accounting for resource inputs, operational structure, and environmental influences. This model provides a critical foundation for identifying negative interactions and inefficiencies, which are further analyzed in subsequent steps through circular economy principles.

Step 2: Present food characteristic indicators based on circular economy principles

The food industry has particular requirements that distinguish it from other sectors [35]. Our research builds upon the circular economy principles from Reichel et al. [36] and is further developed by Baratsas [37]. This study then presents eight key food characteristics aligned with these CE principles, each obeying 1 to 4 indicators, as shown in

Table 1. The food characteristic indicators of circular economy principles.

Category	Indicator		Measurement Approach	Objective
Waste	1a	waste generated	Measured as kg of food residuals and additives in input, converted into a percentage relative to the baseline.	Reduction of food residuals (surplus food, food loss, food waste) and other waste, such as additives
	1b	waste reused	Measured as kg of food residuals reused or valorized in output, converted into a percentage relative to the baseline.	Value maximization of food residuals
Water	2a	water withdrawal	Measured as kl of water consumed in processing, converted into a percentage relative to the baseline.	Reduction of water used
	2b	fresh water discharge	Measured as kl of freshwater discharged, converted into a percentage relative to the baseline.	Reduction of wastewater discharge
	2c	other water discharge	Measured as kl of non-freshwater discharged, converted into a percentage relative to the baseline.
	2d	water recycled or reused	Measured as kl of water recycled or reused, converted into a percentage relative to the baseline.	Efficient use and recovery of water resources
Materials (food)	3a	main materials used	Measured as kg of main food raw materials in input, converted into a percentage relative to the baseline.	Efficient use of main food materials
Procurement	4a	packaging used	Measured as kg of packaging materials consumed in processing, converted into a percentage relative to the baseline.	Reduction or the environmentally friendly adoption of packaging
	4b	auxiliary materials used	Measured as kg of additives and auxiliary materials consumed in processing, converted into a percentage relative to the baseline.	Reduction or the biodegradable adoption of food additives
Energy	5a	energy consumed	Measured as kWh consumed in processing, converted into a percentage relative to the baseline.	Reduction of non-renewable energy used
	5b	energy recovered	Measured as kWh of recovered or renewable energy in processing, converted into a percentage relative to the baseline.	Increase in the share of renewable ones
Emissions	6a	direct emissions	Measured as kgCO2e directly emitted in processing, converted into a percentage relative to the baseline.	Reduction of direct and indirect emissions (e.g., GHG emissions)
	6b	indirect emissions	Measured as kgCO2e indirectly emitted (e.g., electricity use), converted into a percentage relative to the baseline.
	6c	carbon neutralized	Measured as kgCO2e offset or neutralized, converted into a percentage relative to the baseline.	Action adoption of the removal of emissions impact
Pollutants	7a	pollutants generated	Measured as kg of pollutants (e.g., microorganisms, additives) generated in processing, converted into a percentage relative to the baseline.	Reduction of pollutants (e.g., microorganisms, chemical additives)
	7b	pollutants purified	Measured as kg of pollutants treated or purified, converted into a percentage relative to the baseline.	Purification of pollutants
Preservation	8a	quality of food	Measured as product batches maintaining acceptable quality standards, converted into a percentage relative to the baseline.	Enhancement of food preservation through good and stable quality

, which includes their measurement approach and objective. The differences primarily relate to the input and use of natural resources, the nature of pollutants generated, and the durability value. The explanations are the following:

• Raw food materials are critical natural resources. Their availability and quality are essential for process continuity and food security.
• Compared to general industries, the food industry’s concerns about pollutants are more intricate and stringent due to their impact on human health, consumer perceptions, and other considerations.
• Freshness and preservation are central to food durability. Within the CE framework, preservation effectiveness represents a core measure of product lifespan and waste reduction potential.

The research evaluates how each production function in frozen fish processing aligns or conflicts with these CE-based food indicators. The goal is to identify barriers that inhibit the systematic realization of circular practices. A function analysis table is developed to cross-reference each functional component with the CE indicators, enabling a structured assessment and guiding the next problem-solving phase.

Step 3: Determine the function analysis by CE-based food characteristic indicators

Following the function model and the CE food characteristic indicators, as defined in Table 1, this step evaluates each function based on rank and performance level within the production system. This enables the identification of negative interactions and inefficiencies that block circular production goals. The relevant function definitions follow:

Function rank classification

• Basic Function (B): Represents the core objective of a system or product, i.e., the system’s operation that acts on the target and produces a change in its nature to achieve its primary purpose.
• Auxiliary Function (Ax): Supports the realization of the basic function by facilitating necessary interactions between subsystems or components.
• Additional Function (Ad): Reflects the influence of environmental elements (EEs), including physical conditions (e.g., temperature) or adjacent systems (e.g., power supply), on the system.
• Function performance assessment
• Normal Function (N): Functions that meet the system requirements and have a positive (or non-negative) effect on the objectives, ensuring that the system operates according to the design objectives.
• Harmful Function (H): Functions that harm the system or the environment may directly lead to equipment damage and safety issues or cause resource wastage and efficiency degradation.
• Insufficient Function (I): Functions that do not meet the expected requirements, resulting in reduced system efficiency or inability to achieve the goal; when the functional strength is lower than the requirements, it may affect the quality or performance of the product.
• Excessive Function (E): Functions that exceed strength requirements result in wasted resources, additional costs, or reduced system efficiency. Although it may not cause immediate harm, it may affect operational performance in the long term.

To make the analysis more robust for food production, this study integrates the examination of hindering fields or external influences. The evaluation process also incorporates key physical properties of food, such as texture, shape, size, and density, alongside model component effects, notably energy inputs and utilization efficiency. These factors are considered with the eight CE food characteristic indicators (Table 1). Building on the function model from Step 1, this comprehensive analysis assesses each function’s rank (B, Ax, Ad) and performance (N, H, I, E) to generate FA results, which in turn facilitate the identification of negative functions detrimental to circular performance.

For instance, the function carrier “blast freezer” acts on the object “fish material” to achieve the intended function of “freezing.” However, this process simultaneously generates adverse effects related to circular economy performance. Each product unit consumes 11.6 kW of electricity per batch (Indicator 5a) and contributes an average of 6 kgCO₂e of emissions (Indicator 6b). The emission factor used is 0.502 kgCO₂e/kWh, based on data from Taiwan Power Company for the year of our data collection. Due to process variability, approximately 10% of the frozen fish requires rework, which is quantified as rework input waste (Indicator 1a) with a frequency of once every six production cycles. These values establish the baseline performance against which the subsequent circular strategy will be evaluated. From a functional perspective, the blast freezer operates within both the electric and cooling fields. These interactions highlight a composite linkage across energy consumption (5a), emissions (6b), and waste generation (1a) within the CE indicator system. The FA results for this function are shown in

Table 2. The paradigm of the FA result.

Function	Rank	Performance Level		Field
(Blast freezer) Freeze → Fish material	B	E	Energy (5a); Emissions (6b)	Electric; Cooling

.

By applying this analytical approach, the study defines specific improvement objectives to mitigate undesirable functions. These findings set the stage for the next phase, in which systematic innovation tools are applied to develop targeted circular strategies.

2.2. Problem-Solving via Systematic Innovation and Circular Strategies

Building on the previous problem definition, this section presents potential solutions through SFA and 76 standard solutions. These tools help model and resolve the identified functional conflicts. To further ensure the applicability and alignment of these solutions with circular economy objectives, the study introduces QFD to evaluate and prioritize the solutions. The following steps extend the previous analytical process:

Step 4: Propose suitable 76 standard solutions via the presented SFA problem-solving flow

To translate functional inefficiencies into actionable solutions, this step employs Substance-Field Analysis (SFA), a core TRIZ tool designed to formalize and optimize the physical structure of a system. The SFA framework models interactions between “substances” (e.g., materials, equipment) and “fields” (e.g., mechanical, thermal, or electrical energy), forming a Substance-Field Model (SFM) [29]. This model allows the study to diagnose the underlying root causes of production problems extracted from the function analysis in Step 3. In SFA, system components from the function model are reinterpreted as either substances or fields. An ideal SFM typically includes two substances and one field. Compared to Function Analysis (FA), which focuses on a system-level perspective, SFA offers a more granular view by analyzing the interactions between specific system elements and their fields or effects, allowing deeper identification of fundamental inefficiencies. Additionally, SFA is often paired with the 76 standard solutions, a structured TRIZ knowledge base that systematically resolves complex problems by eliminating technical contradictions. These solutions provide general guidance while incorporating constraints or high-level restrictions [38]. Categorized into five groups, 76 standard solutions in categories 1–4 tend to increase system complexity by introducing new materials or field elements, whereas category 5 focuses on simplification to achieve an ideal system state.

The SFA-based problem-solving flow, developed in prior studies to link systemic inefficiencies with applicable 76 standard solutions, has yielded several conceptual frameworks. However, the current flow structure has a flaw in terms of classification or insufficient integrity. As a result, this study investigates the five studies [33,38,39,40,41] and identifies four key deficiencies in existing flows:

• Lack of guidance in selecting specific standard solutions, with an overemphasis on broad categories.
• Insufficient explanation of the 76 standard solutions and their applications.
• Complex and disorganized flow structures make implementation difficult.
• Lack of step-by-step guidance, resulting in ambiguity when selecting appropriate solutions.

To address these limitations, this study integrated common principles from prior research and redefined key terms, including System Problems (SFM type), Interactions, and Enhancement actions:

System Problems (SFM type)

• Potential change: Problems caused by (potential) changes in the functional characteristics of the system.
• System improvement: Problems for enhancement related to system performance and efficiency.
• Detection/Measurement: Problems encountered in the quantitation, monitoring, or evaluation processes of the system.

Interactions

• Minimum change: Actions that are related to minor adjustments, consuming more resources and time.
• Super/sub-system change: Actions related to the environment (super-system) or its internal components (sub-system).
• Inadequate, excessive, or harmful:
• Actions related to insufficient, excessive, and even harmful effects (refer to the performance level of the function).

Enhancement action

• (Dynamic) Rhythm adjustments: Change operation pace, timing, or synchronization to eliminate system performance (insufficient or excessive performance) problems. Operations require a higher degree of interruptibility to adapt to the changes in dynamic processes.
• Ferromagnetism-related action: Enhancing the ferromagnetism material or field to eliminate system performance problems.
• Adding operations: Introducing new operations to improve system performance.
• Replacing operations: Substituting existing operations with alternative ones to resolve system performance problems.
• Modifying operations: Change the existing characteristics of the operation to eliminate system performance problems.
• Transitioning operations: Adjusting operational states or modes to eliminate system performance problems.

These refinements address the limitations found in prior flows and form the basis for the proposed enhanced SFA process, as illustrated in Figure 2. This newly structured flow serves as a practical guideline for effectively applying the 76 standard solutions.

The proposed framework categorizes the standard solutions based on four key discriminants, identified through literature review and observation of standard conditions: (1) the nature of the problem being addressed, (2) the completeness of the problem model, (3) the type of interaction involved, and (4) the direction for enhancing the model.

This research ensures that the problem-solving flow is more systematic, structured, and actionable by considering all categories and sub-items of the 76 standard solutions. The refined SFA problem-solving flow enables a more precise selection of solutions, effectively addressing immediate inefficiencies and underlying causes in frozen food production systems.

Step 5: Comprehensively evaluate and rank the selected standard solutions using QFD

To systematically prioritize the selected TRIZ standard solutions, this step employs Quality Function Deployment (QFD), a structured methodology for translating customer needs into prioritized technical responses [42]. Given their complementary strengths, integrating QFD with TRIZ has been widely endorsed to improve the relevance and efficiency of problem-solving [43,44]. In this study, QFD is applied to rank the selected standard solutions derived from Step 4, identifying the most effective trigger ideas for system improvement. The analysis utilizes the House of Quality (HoQ) framework, in which:

• The technical responses are the selected 76 standard solutions.
• The customer requirements are defined by the identified production problems.
• The prioritization benchmark is based on CE influences, corresponding to the functional inefficiencies (e.g., harmful, insufficient, excessive) linked to the CE food characteristic (Table 1) indicators in Step 3.

As shown in Figure 3, the HoQ matrix incorporates three analytical elements:

• Problem elements—substances and fields extracted from the Su-Field models.
• Selected 76 standard solutions (TRIZ solutions)—screened via the improved SFA problem-solving flow.
• Circular Economy (CE) influences—highlighting deficiencies in CE characteristic indicators and the urgency of intervention.

A relationship matrix is then constructed using a numerical scale of 1 (weak), 3 (moderate), and 9 (strong) [45,46]. This enables a quantitative assessment of how strongly each standard solution addresses specific problems and supports CE-related improvements. The weighted scoring helps rank the solutions in order of priority, ensuring that subsequent actions focus on the most impactful and feasible interventions.

Step 6: Develop the circular economy strategy for frozen food processing

Building on the results of QFD prioritization, this step formulates a circular economy (CE) strategy that integrates the selected high-impact standard solutions with the specific production problems and CE indicator gaps identified earlier. New solution concepts are developed by cross-referencing each ranked standard solution with its associated problem factors and CE targets. Drawing from the detailed explanations and application cases of the standard solutions, we synthesized appropriate design ideas to overcome existing bottlenecks. The resulting CE strategy is then modeled using an updated Su-Field Model (SFM) to illustrate the enhanced structure of interactions and expected outcomes.

To validate the proposed strategy’s effectiveness, we conduct a comparative evaluation of pre- and post-intervention performance based on the CE food characteristic indicators (Table 1). This comparison assesses improvements involving resource efficiency (e.g., reduced energy consumption, minimized waste), environmental impact (e.g., emissions and pollutants), food preservation and quality retention, and so on. Moreover, the strategy is evaluated against the operational conditions of a real production environment, considering available resources such as supply chain partners, production equipment, and technical capabilities. This comprehensive approach ensures that the proposed CE strategy is not only theoretically robust but also applicable and scalable in real-world frozen food processing settings, contributing to long-term sustainability and competitiveness.

3. Case Study and Result Discussion

The frozen fish industry is a vital component of the global food supply chain, known for its high demand, nutritional value, and perishable nature [47]. Improving its sustainability, particularly regarding intensive energy and resource use [8], is crucial. This study applies its proposed methodology to a frozen fish processing case, demonstrating the practical effectiveness of circular transformation in production systems.

3.1. The Case Study Description

Our case study focuses on a small-to-medium-sized enterprise (SME) in Taiwan specializing in seafood import and distribution. The frozen fish product line, which serves both restaurant and retail channels with custom processing, constitutes the company’s largest production segment. Within this line, the freezing process is identified as a critical determinant of product quality, energy efficiency, and sustainability outcomes.

Fish typically contain between 75% and 85% water, making them highly sensitive to freezing conditions [48,49,50]. As shown in Figure 4, the image on the left presents frozen fish from our case study company, while the graph on the right is modified from a typical temperature-water composition phase diagram [51]. This graph illustrates fish’s freezing and thawing behavior, highlighting the glass transition temperature (T_g) at approximately −40 °C, which was identified from the analysis of 125 production data points collected over one year. This finding indicates the optimal freezing temperature for structural integrity. Maintaining a temperature below this critical threshold is essential for minimizing water migration, ice crystal formation, and microbial growth, directly impacting product texture, safety, and shelf life [9]. This also contributes to a longer product shelf life and reduced losses, aligning with circular economy principles such as resource efficiency and value retention [52].

Currently, the company employs a unidirectional controlled blast freezer, operated via a single-host system positioned outside the processing space. A key issue is the interaction between the freezing chamber and ambient temperatures (12–15 °C), which causes significant external fluctuations. These temperature variations lead to inconsistencies in freezing efficiency and product quality by affecting the rate at which the product transitions into the glassy state. This bottleneck highlights the need for an integrated strategy to enhance both operational efficiency and system sustainability. Therefore, this study applies the proposed circular innovation methodology—comprising Function Analysis (FA), Substance-Field Analysis (SFA), 76 standard solutions, and Quality Function Deployment (QFD)—to develop a systematic, knowledge-based transformation strategy for improving the freezing process’s sustainability performance. By targeting both technological and environmental dimensions, the approach aims to improve product quality and reduce resource input simultaneously, which aligns with the broader objectives of circular economy transformation in food systems.

3.2. Challenges to Circular Production in the Frozen Fish Freezing Process

To model the function-oriented system within the frozen fish processing context, this study first identifies key system components, including materials (e.g., fish), equipment (e.g., blast freezer), and environmental factors. Equation (1) structures these elements and delineates their interactions, a total of 7 functions, including:

• Basic function transforms the target object, namely, transitioning fish from semi-frozen to fully frozen.
• Auxiliary functions support this transformation, such as placing fish on an iron plate to facilitate heat transfer.
• Additional functions reflect external or ambient factors that affect system performance but are not directly controlled.

Following the function modeling, this study evaluates the performance of each function using CE food characteristic indicators, as introduced in Table 1. The analysis identifies negative interactions by comparing the current system state against the CE indicators. These evaluations are further structured using the function rank and performance level, distinguishing between normal and suboptimal states.

As shown in Figure 5, the function model classifies system components (white) and super-system interactions (blue), where arrowheads depict the interactions between system components. Negative functions, whether harmful, insufficient, or excessive, are visually identified to highlight improvement objectives.

Three environmental factors are particularly relevant: temperature, water activity, and microorganisms. However, stringent food safety controls within the production environment minimize the effects of water and microorganisms. As such, temperature emerges as the dominant factor affecting system performance.

A notable instance involves the interaction between fluctuating ambient temperatures and the performance of the blast freezer’s host system. Empirical observations were collected from 125 production batches over a year, covering seasonal temperature variations (10–35 °C) and operational variability. The batches were selected to represent different seasons and batch types, including standard and high-demand periods. Temperature and power data were logged for each batch using calibrated instruments installed on the equipment, which were periodically verified for accuracy. All data were cross-checked with production records and maintenance logs, and outliers were reviewed and corrected to ensure quality. Analysis of production records (with a sampling interval of one batch) shows that when external temperatures increased by 2–3 °C, batch processing times extended by approximately 0.2–0.6 h ($∆t$), with an average of about 0.4 h. These additional processing times were converted into equivalent energy consumption, expressed in kWh per batch and normalized per product unit according to the Energy (5a) indicator in Table 1. Therefore, depending on a 2-h baseline batch duration ($t_{base}$), the relative increase in energy demand ($\%∆\mathrm{E}$) with constant average compressor power is calculated using Equation (2).

$$\%∆\mathrm{E}={\displaystyle \frac{∆t}{t_{base}}}\times 100\% =\times 100\%=20\%$$

(2)

While the time–energy relationship is not strictly linear in practice due to system fluctuations and efficiency losses, this approximation reasonably illustrates the additional energy burden under typical conditions. The extended processing by temperature fluctuations not only increases energy demand (5a) but also leads to higher greenhouse gas emissions (6a, 6b), greater material waste (1a, 3a), reduced throughput, and increased operational costs. The corresponding FA results are summarized in

Table 3. The FA result of the frozen fish freezing process.

Function	Rank	Performance Level		Field
(Blast freezer) Freeze → Fish material	B	E	energy consumed (5a); indirect emissions (6b)	Electric; Thermal
(Iron plate) Locate → Blast freezer	Ax	N		None
(Fish material) Locate → Iron plate	Ax	N		None
(Fish material) Change → Rework product	Ad	H	waste generated (1a)	None
(Atmosphere (Temperature)) Retard & Destroy → Whole system/Fish material/Rework product	Ad	H	waste generated (1a); main materials used (3a); energy consumed (5a); direct emissions (6a); indirect emissions (6b); pollutants generated (7a)	Natural (Thermal)
(Water activity) Destroy → Fish material	Ad	N		None
(Microorganisms) Destroy → Fish material	Ad	N		None

.

Based on this model, the production problem can be articulated as follows:

“The blast freezer is adversely affected by external temperature fluctuations, necessitating excessive use of non-renewable energy to stabilize the freezing process. This results in elevated emissions, increased energy consumption, and a higher probability of rework due to product quality inconsistencies.”

To further identify root causes, the function model is transformed into an SFM, as depicted in Figure 6. The analysis reveals that the primary issue stems from the significant temperature differential between the blast freezer’s internal working environment and the uncontrolled external conditions of the host system. This thermal gap disrupts energy efficiency and undermines process stability.

The results indicate that external temperature fluctuations trigger a cascading effect, leading to increased energy use, higher emissions, and greater material waste. These cascading effects compromise eco-efficiency and amplify both production costs and environmental impact. Consequently, the temperature differential is the critical root cause limiting the system’s circular potential, and addressing this challenge is essential for enhancing the freezing process’s sustainability performance. In the next section, we apply the proposed innovation framework to formulate and prioritize circular economy strategies that mitigate the impact of thermal variability and improve overall system performance.

3.3. Circular Economy Strategies for the Frozen Fish Freezing Process

Building on the substance–field model (SFM) developed in the previous section, this study applied the proposed SFA problem-solving flow (Figure 2) to generate circular economy (CE) strategies aimed at enhancing sustainability in frozen fish processing. This SFM (Figure 6) involved four conditions mentioned in Step 4 of Section 2.2 as follows.

(1)

The system involves modifiable components and improvement opportunities.

(2)

The SFM model is complete and interpretable.

(3)

Interactions are inadequate/excessive, harmful, and influenced by super-system variables.

(4)

Strategic improvements involve modification, addition, or reconfiguration of existing system elements.

Given these conditions, the case aligns with three major categories of TRIZ standard solutions: (1) mitigating super-/sub-system effects, (2) reducing harmful interactions, and (3) enhancing (inadequate/excessive) process operations by transition. Following the systematic process outlined in Figure 2, these categories led to the selection of 24 candidate solutions. The specific solutions within these classes were chosen based on their direct relevance to the identified problems and enhancement opportunities. For example, the need to mitigate super-/sub-system effects guided the selection of solutions from Standard Class 3, 2.1, 5.1, 5.3, and 5.4. A complete list of 24 standard solutions is provided in Appendix A.

The QFD matrix (Figure 3) was used to prioritize the 24 selected standard solutions (matrix top rows). This matrix establishes the correlations among the solutions, the functional problem elements derived from the SFM analysis (matrix left columns), and the circular economy (CE) food characteristic indicators (matrix right columns) identified in Table 3. A binary correlation matrix was first constructed to represent the weights of each standard solution, as shown in

Table 4. The binary correlation matrix of the selected 24 standard solutions.

Standard Solutions	1.2.1	1.2.2	1.2.3	1.2.4	1.2.5	2.1.1	2.1.2	……	5.4.1	5.4.2
1.2.1	1	0	0	0	0	1	0	……	0	0
1.2.2	0	1	0	0	1	0	0		0	0
1.2.3	0	0	1	0	0	1	0		0	0
1.2.4	0	0	0	1	0	0	1		1	1
1.2.5	0	1	0	0	1	0	0		0	0
2.1.1	1	0	1	0	0	1	0		0	0
2.1.2	0	0	0	1	0	0	1		0	0
……	……								……
5.4.1	0	0	0	1	0	0	0	……	1	0
5.4.2	0	0	0	1	0	0	0		0	1
Weight	0.194	0.161	0.194	0.355	0.161	0.194	0.194	……	0.323	0.129

. The complete binary correlation matrix is provided in Appendix B.

Each CE indicator was then assigned a weight to quantify its relative importance. For example, the functional element “Blast freezer” is associated with two CE indicators—Energy (5a) and Emissions (6b)—and therefore receives a weight of 2. The correlations in the matrix were further scored using the 1–3–9 rule, which follows Step 5 in Section 2.2. The final priority score (${\mathrm{R}}_j$) for each solution was computed as:

$$R_j = \sum _{i=1}^I(r_{ij}\times w_{CE, i})\times w_{s, i}$$

(3)

where $r_{ij}$ represents the correlation score (1–3–9) between the 24 solutions $j$ and the 5 problem elements $i$ (SFM), $w_{CE,i}$ is the weight of CE indicator $i$, and $w_{s,i}$ is the weight of solution $j$ by Table 4. The QFD evaluation was performed by a panel of five experts: three academic researchers and two production managers, each with over a decade of experience in frozen food production. Individual ratings were first completed independently. The final scores were then determined through a consensus-based process after comparing the initial ratings.

The results are visualized in Figure 7. This structured approach ensures that the most impactful solutions—those addressing both the key functional problems and the most significant circular economy indicators—are systematically identified and prioritized.

From the QFD analysis, five TRIZ solutions were prioritized:

(1)

The useful and harmful effects exist at the same time, but S1 and S2 must be in contact, then increase the field F2 to offset the effect of F1, or obtain an additional useful effect (1.2.4).

(2)

Indirect ways to introduce substances (5.1.1).

(3)

Double Su-Field Model: A poorly controlled system needs to be improved, but you may not change the elements of the existing system (2.1.2).

(4)

Detrimental effects are eliminated by changing S1 or S2 (1.2.2).

System Transition 2: Transition to the Micro-Level (3.2.1)

In this case study, the ambient temperature (S1) directly interacts with the blast freezer (S2) through a thermal field (F1), while S2 processes the fish (S3) using an electro-thermal field (F2) (Figure 4). This configuration forms a classic chain of interactions susceptible to environmental temperature fluctuations. To mitigate this instability, the study proposes transforming the original system into a Double Su-Field Model with an adaptive, three-phase dynamic control mechanism. The innovation lies in redesigning the blast freezer (S2) to incorporate a dual-motor architecture capable of operating under three modes: rapid freezing, temperature maintenance, and storage transition. The dual-motor architecture provides greater power, allowing the equipment to operate more smoothly and significantly increasing production capacity. Simultaneously, a three-phase dynamic controller precisely adjusts energy output for savings. An auxiliary field (F3), representing a freezing cycle, works in parallel with F2 to stabilize the entire process. This dynamic F2-F3 configuration activates the secondary motor only when needed, adapting to environmental variations and reducing energy overload. Figure 8 illustrates the improved SFM.

Existing equipment has a single-motor architecture without three-phase operation. As technology advances, there is a tremendous opportunity to incorporate the design strategies described in this study. New equipment incorporates dual-motor architecture and three-phase operation, which can increase production capacity from 120 kg/hour to 160 kg/hour, an increase of 33.3%, as shown in

Table 5. Evaluation of the effectiveness of the proposed strategy.

Evaluation Criteria	Current Situation ($\mathit{C}$)	Strategy Performance ($\mathit{S}$)	Comprehensive Assessment $\mathbf{(}\mathit{A}\mathbf{=}{\displaystyle \frac{\left|\mathit{S}\mathbf{-}\mathit{C}\right|}{\mathit{C}}}\mathbf{\times }\mathbf{100}\mathbf{\%}\mathbf{)}$
waste generated (1a)	16% rework rate	10% rework rate	Reduce 37.5%
main materials used (3a)	180 kg per processing cycle	180 kg per processing cycle	-
energy consumed (5a)	11.61 kWh	11.18 kWh	Reduce 4%
direct emissions (6a) and indirect emissions (6b)	5.828 kgCO2e	5.612 kgCO2e	Reduce 4%
pollutants generated (7a)	Approaching 0%	Approaching 0%	-
production capacity	120 kg/hour	160 kg/hour	Increase 33.3%

. While this dual-motor mode does temporarily increase instantaneous power demand, it is designed to operate only under specific, high-temperature conditions for a limited time. Based on our data, the increase in instantaneous power is minimal and does not exceed the capacity of the existing power supply system or incur additional peak demand charges.

Building on this, we evaluated the proposed circular strategy using a baseline established from existing technology operations. The evaluation criteria included CE food characteristic indicators, process capacity, and environmental impact. Table 5 summarizes the outcome. We first defined six improved CE indicators from Table 3, measured according to the definitions in Table 1. Our analysis used a standard processing load of 180 kg with a single motor power of 4.3 kW. The proposed strategy, which employs a three-phase controller and a dual-motor architecture, dynamically adapts to temperature fluctuations. Activating a second motor when external temperatures rise boosts production capacity. This dual-motor, dual-freezing cycle maintains a more stable freezing effect, which enhances food quality and reduces the rework rate from 16% to 10% (from one rework in every six processing cycles to one in ten). Additionally, the strategy reduces production time from 1.8–2.7 h to 1.8–2.2 h. The maximum time reduction of 0.5 h directly contributes to savings in energy consumption and emissions.

The results demonstrate that integrating TRIZ-based functional innovation with CE principles enables substantial sustainability improvements. The proposed three-phase freezing strategy led to a 33.3% increase in capacity without added material inputs, a 37.5% reduction in waste (1a), and a 4% decrease in both energy usage (5a) and emissions (6a, 6b). These outcomes underscore the innovation’s potential to support circular transformation in temperature-sensitive food production systems. This improvement is primarily due to the dual-motor architecture and a dynamic control system. In high-temperature environments, the second motor can be activated to quickly handle thermal loads, shortening the processing time per batch. This optimized process significantly increases system throughput, translating directly into a gain in capacity without requiring additional material inputs.

Implementation is expected to require an initial investment of approximately NT$1.2 million, and the short-term operational cost may rise by 22–30%. However, the strategy offers a projected payback period of 1.2 to 1.5 years and is expected to increase long-term production efficiency by 5–8%.

4. Conclusions

This study introduces an innovation-driven framework that enables the circular transformation of frozen food processing by directly addressing temperature-related inefficiencies—an often overlooked yet critical barrier to sustainable food production. Our integrated approach, leveraging Function Analysis (FA), Substance-Field Analysis (SFA), the TRIZ 76 Standard Solutions, and Quality Function Deployment (QFD), systematically uncovers root causes of inefficiency and guides the development of pivotal, inventive strategies for improvement.

A core innovation of this research lies in its dynamic mapping of food-specific circular economy (CE) characteristics to production system functions. This novel mapping allows for precisely identifying complex system interactions, particularly those arising from challenging temperature gradients. The practical efficacy of the approach was validated in a frozen fish processing case study, demonstrating its potential to achieve tangible sustainability improvements in production capacity, waste reduction, energy efficiency, and emissions reduction—all without additional resource inputs. The strengths of this study include:

• A structured and holistic method that blends inventive problem-solving (TRIZ) with CE principles and stakeholder-driven prioritization (QFD).
• The dynamic mapping of food-specific CE characteristics to production functions enables the precise identification of complex system interactions.
• Validation through a real-world industrial case highlighting practical viability and scalability potential.

However, several limitations should be noted:

• The current framework has been tested primarily on frozen fish processing, and its applicability to other food sectors requires further investigation.
• The framework involves specialized knowledge of TRIZ tools, which may pose adoption challenges for practitioners unfamiliar with this method.
• Integration of broader environmental, economic, and social sustainability dimensions remains an area for future development.

In summary, this research fundamentally advances the integration of inventive problem-solving techniques within the circular design of food production systems. By embedding TRIZ logic and CE principles into a structured, holistic design methodology, it offers a repeatable, adaptable pathway for accelerating sustainability in practice. Future research will focus on extending the applicability and scalability of this robust framework across various food industry sectors. Emphasis will be placed on refining the classification of circular pathways by thoroughly incorporating environmental, economic, and social dimensions, thereby supporting the evolution of a more resilient, efficient, and sustainable food sector.