The Optimization of Cyclic Links of Live Pig-Industry Chain Based on Circular Economics

Xing Liu, Xu XiaoSustainability 2015 · DOI

TL;DR

This paper tackles the significant environmental challenges posed by the pig-industry chain by proposing a circular economy model. It develops a multi-objective optimization framework to determine optimal wastewater and waste treatment capacities, aiming to maximize resource reuse and minimize costs under uncertain market demand, using a hybrid neural network and genetic algorithm approach.

What Problem Does This Paper Address?

The live pig-industry chain, despite its economic importance, is a major source of pollution due to manure, sewage, and noise. Traditional linear economic models exacerbate this by focusing on resource consumption and waste disposal. This paper addresses the critical need to transition to a circular economy model for the pig industry to mitigate environmental damage, improve resource utilization, and achieve sustainable development. Specifically, it focuses on optimizing the "cyclic links" within the chain, which involves determining the most efficient pathways for material reuse and designing appropriate capacities for wastewater and waste treatment facilities to handle the byproducts of pig farming and slaughtering. The challenge is compounded by the inherent uncertainty in market demand, which directly impacts the required processing capacities.

Methods

The core of this paper is the development of a multi-objective uncertainty-optimization model. This model aims to simultaneously achieve several goals: maximizing resource reuse efficiency (specifically recycled water, manure, and feed) and minimizing construction costs for wastewater treatment and waste disposal facilities. The "uncertainty" aspect is handled by incorporating probabilities and confidence levels into the constraints, reflecting the unpredictable nature of market demand.

To solve this complex optimization problem, the authors employ a hybrid intelligent algorithm combining:

  • Neural Networks: Used to approximate the uncertainty functions within the model. This allows for the estimation of probabilities and the handling of complex relationships between variables.
  • Genetic Algorithms (GA): A global optimization technique used to find the optimal solutions for the multi-objective problem. The paper describes a "double-layer" genetic algorithm, suggesting an inner loop for optimizing resource distribution and an outer loop for optimizing treatment capacities.

The process involves:

  1. Defining the uncertainty functions based on system constraints.
  2. Generating input-output samples for these functions through random simulation.
  3. Training neural networks to approximate these functions.
  4. Using a genetic algorithm to search for optimal solutions for treatment capacities and resource allocation.

Datasets

The paper does not introduce new benchmark datasets. Instead, it relies on historical data to determine:

  • Market demand distributions for agricultural products, pig-slaughtering products, and other breeding industries.
  • Relevant parameters for the pig-industry chain, such as water demand, feed coefficients, waste production rates, and conversion efficiencies.

These parameters are then used to define the model and its constraints, and the case study utilizes specific values derived from this historical data.

Key Insights and Ideas

  • Quantifying Circularity: The paper moves beyond qualitative discussions of circular economy to provide a quantitative framework for optimizing its implementation in a specific industry.
  • Uncertainty is Key: Recognizing and modeling market demand uncertainty is crucial for designing robust and efficient circular systems, preventing both under-capacity (pollution) and over-capacity (waste).
  • Integrated Optimization: The model effectively links wastewater treatment, waste disposal, and resource recycling, demonstrating that these components are interdependent and must be optimized holistically.
  • Hybrid Intelligence for Complex Problems: The combination of neural networks and genetic algorithms proves effective in tackling the non-linear and probabilistic nature of the optimization problem.
  • Cost vs. Efficiency Trade-off: The multi-objective approach explicitly addresses the inherent trade-off between environmental performance (resource reuse) and economic considerations (construction costs).

Key Results

The case study, based on parameters derived from historical data for a specific region, yielded the following optimal capacities and resource allocations:

  • Wastewater Treatment Capacity (Q1): 591.73 m³
  • Waste Disposal Capacity (Q2): 23,155.90 kg

The optimized distribution of treated recycled water and waste products is also detailed:

  • Recycled Water Allocation:
    • Planting industry: 186.34 m³
    • Pig-breeding industry: 71.12 m³
    • Pig-slaughtering industry: 207.25 m³
    • Other livestock-breeding industries: 45.33 m³
  • Recycled Waste Allocation:
    • Manure for planting industry: 83.62 kg
    • Feed for other breeding industries: 9639.10 kg

These results demonstrate the model's ability to determine specific, actionable capacities and allocation strategies that balance resource utilization and cost.

Limitations and Future Work

The authors acknowledge several limitations and suggest avenues for future research:

  • Time-Variant Characteristics: The current model does not fully account for the dynamic changes in pig weights throughout the breeding process, which can affect waste production and feed requirements.
  • Transport Costs: Variations in transport costs for distributing cyclic materials between different stages of the chain are not explicitly modeled.
  • Matching Time Demands: The precise temporal matching between wastewater and waste processing times and their respective demands is simplified.
  • Further Factors: The paper notes that other factors, such as spatial location and capacity variations within the chain, and dynamic changes in practical paths, could further optimize the system. Future work will focus on incorporating these dynamic and spatial elements to achieve more practicable and effective results.

Summary updated

Abstract

To reduce waste and wastewater pollution and to improve the utilization rate of resources in the pig-industry chain, a circular economy of the chain can be developed. The key to constructing the circular economic system of the pig-industry chain is to determine the path of the cyclic materials and to design reasonable waste- and wastewater-treatment capacities. This paper focuses on the treatment and recycling of wastewater in the pig-industry chain and the treatment and recycling of waste into manure and feed. After giving the two circular paths, the paper proposes a multi-objective uncertainty-optimization model for the cyclic links of the pig-industry chain with the highest resource-reuse efficiency and the lowest construction cost based on the uncertainty of market demand. Using a combination of the neural network and genetic algorithm method for designing the solution process for the model, the paper finally introduces the determination methods of relevant parameters and verifies the feasibility and effectiveness of the model through a case study.
BibTeX
@inproceedings{liu2015optimization,
  author = {Xing Liu and Xu Xiao},
  title = {{The Optimization of Cyclic Links of Live Pig-Industry Chain Based on Circular Economics}},
  booktitle = {Sustainability},
  year = {2015},
  doi = {10.3390/su8010026},
}
Figure