TL;DR
This paper tackles the challenge of optimizing sustainable insect production by developing a multi-objective mathematical model that balances economic viability, environmental impact, and societal benefits. Using the NSGA-II algorithm, they demonstrate a promising approach to finding trade-offs for designing more sustainable insect farming operations.
What Problem Does This Paper Address?
The burgeoning field of insect farming, while promising for sustainable protein production, faces significant hurdles in optimizing its chains. Current industrial insect farming is relatively new, and there's a lack of comprehensive studies on how to best balance its various objectives. This paper addresses this gap by proposing a framework to simultaneously optimize for three key, often conflicting, goals: maximizing economic profit, minimizing environmental footprint, and enhancing societal benefits. This is crucial because simply maximizing one objective (like profit) can negatively impact the others, leading to unsustainable practices.
Methods
The authors employ a multi-objective evolutionary algorithm, specifically NSGA-II (Non-dominated Sorting Genetic Algorithm II), a well-established technique for finding a set of optimal solutions (Pareto front) when dealing with multiple conflicting objectives. They developed a mathematical model that encapsulates three objective functions:
- Economic Aspect (F₁): Maximizing operating profit, considering factors like product price, feed conversion efficiency, and labor costs.
- Social Aspect (F₂): A weighted combination of Fair Wage Potential (FWP) and Labor Safety (LS), aiming to ensure fair compensation and safe working conditions.
- Environmental Aspect (F₃): Minimizing insect frass (excrement), which is linked to transportation emissions and disposal costs.
The model represents an "individual" (a potential solution) as a genome containing variables for facility scale, labor, feed input, wages, and safety equipment usage. NSGA-II is then used to evolve a population of these individuals towards an optimal Pareto front.
Datasets
Not discussed in the paper. The authors mention that parameters are based on interviews with insect farmers and experts, and data from existing literature, particularly concerning the Netherlands for parameters like IEF, RWT, MLW, and CWT.
Key Insights and Ideas
- Conflicting Objectives are Real: The paper clearly illustrates that economic, social, and environmental objectives in insect farming are not always aligned. For instance, increasing production (economic) can lead to more frass (environmental), and optimizing for profit might conflict with fair wages.
- Scale Matters Significantly: The results suggest that insect production might be most viable at either very small or very large scales, with a discontinuity in the Pareto front for intermediate scales. This hints at potential economies of scale or specific operational efficiencies at these extremes.
- Social and Environmental Trade-offs are Less Antagonistic: Interestingly, the study indicates that social and environmental aspects might not be as directly opposed as initially thought, with many solutions achieving near-optimal social values across a range of environmental impacts.
- Frass as a Proxy: While frass is used as a key environmental indicator, the authors acknowledge this is a simplification. A more comprehensive Life Cycle Assessment (LCA) would provide a richer environmental picture.
Key Results
The experimental evaluation produced approximate Pareto fronts, visualized in 2D projections (Figures 2, 3, 4, 5). These plots reveal trade-offs:
- Economic vs. Environmental: A steep linear relationship exists for some solutions, where improving economic outputs significantly worsens environmental impacts. However, after a discontinuity, the trade-off becomes softer.
- Economic vs. Social: Small-scale facilities tend to achieve better social outcomes more easily. Larger facilities face more challenges in balancing economic gains with social aspects like fair wages and safety.
- Social vs. Environmental: A wide range of environmental impacts can be achieved with near-optimal social values, suggesting flexibility in this dimension.
Table 2 provides specific fitness values for selected points on the Pareto front, showcasing example trade-offs between the three objectives for different facility scales.
Limitations and Future Work
The authors acknowledge several limitations:
- Simplified Models: The current models are a first step and do not capture all complexities of insect production chains. For example, scale effects on waste disposal efficiency are ignored, and the environmental impact of large-scale facilities might be overestimated.
- Limited Scope: The current work focuses on the Netherlands and does not include factors like transportation of goods to and from facilities or variations across different European countries.
- Frass as a Single Metric: Using only frass as an environmental indicator is a simplification.
Future work proposed includes:
- Broader Insect and Location Scope: Incorporating different insect species and various production locations, accounting for country-specific parameters like labor and energy costs.
- More Comprehensive Metrics: Integrating complete LCA for environmental impact and more accurate estimates of capital/operational expenditures for economic viability.
- Including Transportation: Adding parts of the supply chain such as transportation of goods.
Summary updated