Elsevier

Applied Energy

Volume 130, 1 October 2014, Pages 289-304
Applied Energy

A network design model for biomass to energy supply chains with anaerobic digestion systems

https://doi.org/10.1016/j.apenergy.2014.05.043Get rights and content

Highlights

  • A MILP model is developed to design and operate bioenergy supply chains.

  • Anaerobic digestion is considered as biomass to energy conversion process.

  • A real-world application is performed with real data in İzmir, Turkey.

  • A reasonable level of profit can be made by the biomass to energy investment in İzmir.

  • An acceptable payback period of 4.98 years is obtained for the investment.

Abstract

Development and implementation of renewable energy systems, as a part of the solution to the worldwide increasing energy consumption, have been considered as emerging areas to offer an alternative to the traditional energy systems with limited fossil fuel resources and to challenge environmental problems caused by them. Biomass is one of the alternative energy resources and agricultural, animal and industrial organic wastes can be treated as biomass feedstock in biomass to energy conversion systems. This study aims to develop an effective supply chain network design model for the production of biogas through anaerobic digestion of biomass. In this regard, a mixed integer linear programming model is developed to determine the most appropriate locations for the biogas plants and biomass storages. Besides the strategic decisions such as determining the numbers, capacities and locations of biogas plants and biomass storages, the biomass supply and product distribution decisions can also be made by this model. Mainly, waste biomass is considered as feedstock to be digested in anaerobic digestion facilities. To explore the viability of the proposed model, computational experiments are performed on a real-world problem. Additionally, a sensitivity analysis is performed to account for the uncertainties in the input data to the decision problem.

Introduction

Decision making for design, operation and management of bioenergy supply chains are increasingly gaining importance in recent years parallel with the rising interest in renewable energy sources. Strategic, tactical and operational level decisions about the locations and capacities of conversion plants and storages, logistics issues and transportation network, feedstock procurement, handling and distribution of process residue, and tactical operation schedules should be made efficiently to obtain robust and cost effective supply chain configurations. Many decisions in such a supply chain involve tradeoffs. For instance, locating the facilities close to demand points of the products will reduce the transportation cost of these products, but might increase the biomass transportation cost if the facilities are far away from the biomass supply regions. Due to the complex tradeoffs involved, various competing supply chain network design decisions cannot be made independently. Therefore, comprehensive management and optimization of all of the individual components along the entire supply chain is essential to facilitate the economical, environmental and social benefits of bioenergy systems.

Decision making in bioenergy supply chain problems requires getting a sound grasp of the supply chain structure and selection of the suitable methodologies as in many complex planning problems. Before the use or development of the methods, it is important to investigate the current literature in the field to prevent research overlaps. Gold and Seuring [1] presented a literature review of papers from 2000 to 2009 which deal with bioenergy production, logistics and supply chain management, and sustainability issues. An et al. [2] provided a literature review of researches on decision making in biofuel supply chains. They categorized the surveyed studies according to the decision level they include as well as level in the supply chain. Iakovou et al. [3] proposed a critical synthesis of the state of the art literature about design and management of waste biomass supply chains. They mentioned components, activities and characteristics of the supply chain as well as types of biomass to energy conversion technologies.

The state of the art analysis also provides a better comprehension of the technical details of conversion processes and characteristics of biomass sources as well as the supply chain structure. Anaerobic digestion is a well-known and efficient process that converts organic feedstock into biogas by biologic reactions in the absence of oxygen. Donoso-Bravo et al. [4] presented an overview of the modeling procedures for anaerobic digestion processes focusing on mathematical modelling, methods for parameter estimation and optimization, and model validation. Ariunbaatar et al. [5] reviewed mechanical, thermal, chemical and biological pretreatment methods for anaerobic digestion of organic solid waste as well as combination of various pretreatment methods. They compared the methods and evaluated the feasibility of application. Srirangan et al. [6] focused on clean energy production from biomass resources. After defining the first, second and third generation biomass feedstocks, they mentioned biomass to energy conversion routes and various types of biofuels as clean energy carriers. Browne and Murphy [7] proposed a study to assess the resource for biomethane production with a focus on food waste. They dwelt on the technical issues in conversion process such as preparation of food waste and experimental setup as well as biomethane potential of the food waste. Prajapati et al. [8] proposed a study that discusses and compiles main issues about procurement and anaerobic digestion of algal biomass such as growth requirements for algae cultivation, harvesting and cultivation methods, digestibility and biogas potential analyses, limitations of the process. They especially focused on wastewater utilization as the nutrient and waste gases as the CO2 source for algal biomass production.

Considerable research has been conducted on developing mathematical models to optimize the design and operation of various configurations of biomass to energy supply chains. Among the mathematical modeling approaches, mixed integer linear programming (MILP) has been widely utilized to design and operate bioenergy supply chains. It is a powerful tool for such problems because of its modeling capability and the availability of efficient solvers. One of the advantages of MILP approach is that it provides a general framework for modeling a large variety of problems. However, the major difficulty lies in the computational expenses that may be involved in solving large scale problems, which is due to the computational complexity of MILP problems.

Papapostolou et al. [9] developed a mathematical model to identify the best solutions for the optimal design and operation of biofuel supply chains that takes into account both technical and economic parameters affecting the performance of the supply chain. Zhang et al. [10] proposed a MILP model to determine optimal bioethanol supply chain/logistics decisions. Biomass cultivation sites are selected, biomass inventory level is determined, location and capacities of biorefineries and preprocessing plants are selected, and production/transportation volume of bioethanol is determined by the model. Zhu and Yao [11] presented a MILP model to design the entire supply chain and logistics system for the biomass to energy industry which incorporates multiple types of biomass and products. The model integrated strategic and tactical level decisions. Akgul et al. [12] developed a static MILP framework for strategic design of bioethanol supply chain network to determine locations and scales of biofuel production facilities, biomass cultivation and biofuel production rates, flow of biomass and biofuel between the components of supply chain, transportation modes of delivery for biomass and biofuel. Giarola et al. [13] presented a MILP framework to optimize the environmental and financial performances of corn grain and stover based bioethanol supply chains simultaneously. Dal-Mas et al. [14] focused on biomass based ethanol supply chain design under uncertainty conditions. A MILP is developed to model entire corn to ethanol supply chain determining biomass cultivation site locations, capacities and locations of ethanol production facilities and logistic issues. Kim et al. [15] formulated a MILP model to decide the optimal number, locations and sizes of various types of biofuel processing plants as well as fuel conversion technologies and logistics issues such as transportation amounts of biomass intermediate products and final products between supply areas to conversion plants and between conversion plants to final markets. Kim et al. [16] improved Kim et al. [15]’s model by considering the uncertainties in the model parameters. They formulated a general MILP model for a simple biorefinery network structure for single and multiple design scenarios.

Simulation and heuristic based methodologies have also been used by a number of researchers. Heuristic methods are generally used for solving large scale and/or non-linear programming models that generally look for not optimal but good enough solutions in relatively reduced times. Ebadian et al. [17] proposed a methodology that combines MILP and simulation for determining network layout and operational design of agricultural biomass supply chain. Once the design of the supply chain is determined, operational decisions are made by the simulation model. Sokhansanj et al. [18] developed a dynamic integrated biomass supply analysis and logistics model (IBSAL) to simulate the flow of biomass from field to biorefinery. IBSAL is a modeling approach to plan resources and estimate costs, energy use and GHG emissions for different collection and transportation systems. Ebadian et al. [19] proposed a stochastic simulation model based on IBSAL to provide optimal operational plan and schedule for logistics operations to supply a mixture of agricultural biomass to a proposed cellulosic ethanol plant considering variable data on weather conditions, harvest schedule and yields. Wang et al. [20] utilized simulation to economically analyze bioethanol production from waste papers. Bioethanol supply chain is modeled to compare the selling price of bioethanol produced from waste paper with petrol price. Gómez-González et al. [21] developed a hybrid heuristic approach to find the best location and size of biomass fuelled electricity generation facilities. To this aim, Particle Swarm Optimization (PSO) is used to search a range of location combinations in the distribution network and Optimal Power Flow (OPF) is utilized to define available capacity for each combination. In the study a novel discrete PSO is proposed, namely Jumping Frog PSO (JFPSO) and integrated this method with OPF. López et al. [22] applied and compared several metaheuristic techniques to optimize the location and biomass supply area of biomass based power plants. For this purpose two trajectory (Simulated Annealing and Tabu Search) and two population-based (Genetic Algorithms and PSO) methods are applied. After explaining the four considered metaheuristic methods, a new PSO algorithm is proposed, all above mentioned methods are applied and comparative results are given.

Modeling and optimization approaches for biomass to energy supply chain network design of increasing scope and sophistication have been devised recently. However, network design models for the bioenergy supply chains including anaerobic digestion facilities have not been dealt with in the previous research although it is one of the most efficient and environment friendly energy production systems. In addition, the vast majority of the studies on bioenergy supply chain design considers only energy crops as biomass resource. Considering these facts and research gaps, this study presents a MILP model to obtain optimal configuration of biomass to energy supply chains including anaerobic digestion systems and designs a multi-commodity distribution network.

In this study, waste biomass is considered as the main source to be converted to energy by anaerobic digestion. As improper disposal or storage of organic wastes causes pollution in underground and surface waters and environmental problems that threat human health, constructing waste biomass to energy conversion plants is vital in decreasing environmental problems besides gaining economical benefits by the energy production. To explore the viability of the proposed model, computational experiments are performed on a real-world problem in İzmir, Turkey. This study is the first attempt to design and analyze a comprehensive biomass to energy supply chain network in Turkey.

The supply chain network considered in this study (see Fig. 1) includes the following elements.

  • A set of biomass sites where biomass materials are available.

  • A set of candidate sites for the location of biomass to energy conversion plants with various capacity options.

  • A set of candidate sites for the location of storages with a pre specified maximum capacity level.

The main purpose of this study is to make strategic, tactical and operational decisions to optimize biomass to energy supply chain networks considering economic and environmental criteria. The major activities performed in the supply chain that incorporated in the model are as follows:

  • Biomass collection and transportation from the biomass supply regions to the storages.

  • Biomass transportation from storages to bioenergy plants.

  • Biogas production which refers to the transformation of biomass into biogas through anaerobic digestion process.

  • Biogas to energy conversion through cogeneration.

  • Transportation of fertilizer (residue of digestion) to biomass supply regions.

The remainder of this study is organized as follows: Section 2 presents the problem in detail, while the proposed model is presented in Section 3. To explore the viability of the proposed model, Section 4 presents computational experiments on a real-world problem. Section 5 concludes the study.

Section snippets

Problem statement

Biomass in the form of animal manure and energy crops is considered as feedstock in the proposed model. Animal manure can be taken from available farms and energy crops can be collected from marginal lands not used for other agriculture purposes. Then, the biomass will be shipped to the storages using trucks and tankers. After a storage period, biomass is shipped to the biomass to energy conversion plants. The biomass is converted into biogas in the plants through anaerobic digestion process.

The proposed model

In this section, an optimization model is proposed for the multi-biomass and multi-product supply chain system. The proposed model is formulated as a MILP model, and it aims to maximize the total profit by prescribing an optimal supply chain design. The model can be adapted to various practical problems in different regions with different biomass types. The notation, objective function and constraints of the proposed model are presented in the following.

Computational experiments

To explore the viability of the proposed model, computational experiments are performed in this section on a real-world problem. In this regard, we aim to design a biomass to energy supply chain network in İzmir, which is the third largest city in Turkey with a population of around 4 million. As agriculture and animal husbandry are among the most common economic activities in İzmir, there exists a large waste biomass resource which can be used as feedstock in biomass to energy conversion

Conclusions

The objective of this study is to develop an effective supply chain network design model for the production of biogas through anaerobic digestion of biomass. To this aim, a MILP model is developed to determine the most appropriate locations for the biogas plants and biomass storages. Mainly, waste biomass is considered as feedstock to be digested in anaerobic digestion facilities. The objective of the proposed model is to maximize the profit of the entire supply chain. Besides the strategic

Cited by (0)

View full text