Food industry plays an important role in providing basics and necessities for supporting various human activities and behaviors (Cooper and Ellram, 1993). Once harvested or produced, the food should be stored, delivered, and retailed so that they could reach to the final customers by due date. It was reported that about one-third of the produced food has been abandoned or wasted yearly (approximately 1.3 billion tons) (Manning et al., 2006). Two-third of the wasted food (about 1 billion tons) is occurred in supply chain like harvesting, shipping and storage (Fritz and Schiefer, 2008). Take fruit and vegetables for example, such perishable food was wasted by 492 million tons worldwide in 2011 due to the inefficient and ineffective food supply chain management (FSCM) (Gustavsson et al., 2011). Therefore, FSCM is significant to save our food.
Food Supply chain graph :
2.Supply chain management for food industry
2.Supply chain management for food industry
A framework for FSCM is a basis for manufacturing, processing, and transforming raw materials and semi-finished products coming from major activities such as forestry, agriculture, zootechnics, finishing, and so on (Dubey et al., 2017). In order to identify the relationships among different items, interpretive structural modeling (ISM) was used to establish a hierarchical framework (Faisal and Talib, 2016). This framework helps users to understand the interactions among logistics operators in a food supply chain. ISM-enabled framework was also used to support risk management in identifying and interpreting interdependences among food supply chain risks at different levels such as first-tier supplier, third-party logistics (3PL), etc. (Colin et al., 2011). It is observed that this framework was proven as a useful method to structure risks in FSCM through a step-by-step process on several manufacturing stages. Information plays an important role in making FSCM more efficient. In order to assess the information risks management, an ISM based framework was proposed by twining graph theory to quantify information risks and ISM to understand the interrelationships in FSCM (Nishat Faisal et al., 2007). As the global FSCM is emerging with international collaborations, ISM-enabled framework confines to explain causal relationships or transitive links among various involved parties. A total interpretive structural modeling was then introduced to analyze some enablers and barriers of FSCM (Shibin et al., 2016). In this paper, ten enablers and eight barriers are examined by separate frameworks to further understand the interactions within a dynamic era of globalization FSCM.
Value chains play a critical role in FSCM to benefit the producers and consumers. Stevenson and Pirog (2008)introduced a value chain framework for strategic alliances between food production, processing and distribution which seek to create more value in the supply chain. The proposed framework concerns about food supply chain economic performance that correspond to the organization, structure, and practices of a whole supply chain. Food traceability has been widely used in the last few decades with large number applications. However, frameworks for a general or common implementation are scarcely reported. To label whether a framework with respect to food traceability application, Karlsen et al. (2013) observed that with a common framework, traceability is prone to be similar and implementation processes are more goal-oriented and efficient. Thus, Regattieri et al. (2007) presented a general framework and used experimental evidence to analyze legal and regulatory aspects on food traceability. They designed an effective traceability system architecture to analyze assessment criteria from alphanumerical codes, bar codes, and radio frequency identification (RFID). By integrating alphanumerical codes and RFID technology, the framework has been applied for both cheese producers and consumers.
Currently, coordination in the food supply chain from production to consumption is significant to ensure the safety and quality of various food. Take agri-food supply chain for example, Hobbs and Young (2000) depicted a conceptual framework to achieve closer vertical collaboration in FSCM using of contracting approaches. This work has critical impacts on transaction cost economics by developing a closer vertical coordination. In an international food supply chain, Folkerts and Koehorst (1997) talked about a framework which integrates the chain reversal and chain management model to make vertical coordination. In their framework, an analytical service designed particularly for benchmarking food supply chain projects is used so that an interconnected system of high performance and effectiveness are achieved as an integrated supply chain. Facing a global FSCM, strategic decision-making is important since the profitability of an entire chain could be increased by the holistic efforts from an efficient framework. To this end, Georgiadis et al. (2005) presented a system dynamics modeling framework for the FSCM. In this framework, end-users are able to determine the optimal network configuration, inventory management policy, supply chain integration, as well as outsourcing and procurement strategies. Collaboration is becoming more of a necessity than an option despite some barriers which deteriorate coordination among enterprises in food industry all over the world. Doukidis et al. (2007) provided a framework to analyze supply chain collaboration in order to explore a conceptual landmark in agri-food industry for further empirical research. It is observed that, from this framework, supply chain collaboration is of critical importance and some constraints such as time and uncertainties arise due to the nature of agri-food industry.
It is no debate that IT systems are essential for FSCM where so many things can go wrong such as trucks, food suppliers, data entry, etc. This section takes the traceability and decision-making systems for FSCM as examples to review the state-of-the-art situations that are useful for practitioners when they are implementing IT-based solutions.
3.1Traceability systems
Traceability of a food refers to a data trail which follows the food physical trial through various statuses (Smith et al., 2005). As earlier as two decades ago, US food industry has developed, implemented, and maintained traceability systems to improve FSCM, differentiate foods with subtle quality attributes, and facilitate tracking for food safety (Golan et al., 2004a). Some systems deeply track food from retailer back to the sources like farm and some only focus on key points in a supply chain. Some traceability systems only collect data for tracking foods to the minute of production or logistics trajectory, while others track only cursory information like in a large geographical area (Dickinson and Bailey, 2002).
This section analyzes total 19 key papers published from 2003 to 2017. Table II presents a categorized analysis in terms of tracing objects, technology, district, and features.
From Table II, it could be observed that food traceability is paid much attention from EU where people do care more about the food safety and quality. Associated technologies are developing fast so that cutting-edge techniques are widely used for various food tracing and tracking. Take RFID for example, 73.68 percent of the reviewed papers adopt this Auto-ID technology for food traceability. Moreover, agri-foods are placed special attention to trace and track because as the most important perishable products, their freshness and quality are eyed by the consumers.
3.2Decision-making systems
Besides the traceability systems in FSCM, other decision-makings such as integration/collaboration, planning/scheduling, fleet management, and WMS are also widely used in food industry. This section presents a review of total 26 papers which are related to the above topics. Table III reveals these papers from 2005 to 2007 with specific decisions, countries/area (identified by the corresponding author), used technologies, and features.
We selected two typical publications in each year for forming Table III from which several observations could be achieved. First, European countries are prone to be more use of systems to assist decision-makings in FSCM. Second, systems used in earlier stage are based on internet solutions. Currently, model-based systems using advanced technologies are widely reported in FSCM decision-makings. Third, focuses of decision-making shift from supply chain integration in earlier years to sustainable and specific problem solving cases in recent years.
4.1Reported cases
Case studies from implementing various IT systems in FSCM are significant to get some lessons and insights, which are meaningful for industry practitioners and research academia. This section reports several cases using different systems for facilitating their operations or decision-makings in food supply chain from 2007 to 2017. They are categorized in the following Table IV which includes key information like company name, district, system, and improvement.
From the reported cases, it could be observed that, European countries have much more successful cases on using various IT support systems in FSCM. While, cases from Australia, China, etc. are scarcely presented. Another interesting finding is that before 2010, IT systems are used for optimization or supply chain coordination decision-makings. However, currently, companies are more concentrating on the sustainability and environmental performance in the food supply chain. For example, environmental influences like CO2 emissions and waste reduction are widely considered.
4.2Data-driven implementations
Data, usually used for decision-makings, have been considered in FSCM for various purposes. Data-driven implementation in FSCM is categorized into two dimensions in this paper. First is the simulation-based modeling which focuses on adopting different data for FSCM optimization or decision-making. The other is data collection from practical implementations for supporting IT systems for various purposes such as traceability, risk assessment, and so on.
For simulation-based modeling, studies mainly focus on establishing various simulation models which adopt different types of data such as product quality, customer demand for different decision-makings and predictions. In order to meet increasing demand on food attributes such as integrity and diversity, Vorst et al. (2009) proposed a simulation model which is based on an integrated approach to foresee food quality and sustainability issues. This model enables effective and efficient decision support on food supply chain design. FSCM is becoming more complex and dynamic due to the food proliferation to meet diversifying and globalizing markets. To make a transparent food supply chain, Trienekens et al. (2012) simulated typical dynamics like demand, environmental impacts, and social aspects to enhance the information sharing and exchanging. It is found that food supply chain actors should provide differentiated information to meet the dynamic and diversified demands for transparency information. As a wide application of Auto-ID technology for tracking and tracing various items (Zhong, Dai, Qu, Hu and Huang, 2013; Zhong, Li, Pang, Pan, Qu and Huang, 2013; Qiu et al., 2014; Guo et al., 2015; Scherhaufl et al., 2015), traceability data plays an important role in supporting FSCM. Folinas et al. (2006) introduced a model which uses the traceability data for simulating the act guideline for all food entities in a supply chain. The assessment of information underlines that traceability data enabled by information flow is significant for various involved parties in food supply chain to ensure food safety. Wong et al. (2011) used a model to evaluate the postponement as an option to strengthen food supply chain performance in a soluble coffee manufacturer. The simulation model shows that cost savings including reduction of cycle stock are obtained by delaying the labeling and packaging processes. Bajželj et al. (2014)simulated the food demand to examine the impacts of food supply chain on climate mitigation. This paper proposes a transparent and data-driven model for showing that improved diets and reduced food waste are critical to deliver emissions reductions. Trkman et al. (2010) used a structural equation model based on data from 130 companies worldwide to examine the relationship between analytical capabilities in FSCM. It is observed that the information support is stronger than the effect of business process orientation in food supply chain. Data-driven model was also proposed by developing a measure of the captured business external and internal data for food productivity, and supply chain value (Brynjolfsson et al., 2011). This paper obtains 179 firms’ data from USA where 5-6 percent increase in their output and productivity by using IT solutions. Low and Vogel (2011) used a national representative data on local food market to evaluate the food supply chain where small and medium-sized farms dominate the market. This paper finds that direct-to-consumer sales of food are greatly affected by climate and topography which favor perishable food production. Akhtar et al. (2016) presented a model by using data collected from agri-food supply chains to examine adaptive leadership performance in FSCM. This paper thus depicts that how global food supply chain leaders can use data-driven approach to create financial and non-financial sustainability. Hasuike et al. (2014)demonstrated a model to simulate uncertain crop productions and consumers’ demands so as to optimize the food supply chain profit. This simulation model is based on stochastic programming that accommodates surplus foods among stores in a local area. Manning et al. (2016) used a quantitative benchmarking model to drive sustainability in food supply chain. Li and Wang (2015) based on networked sensor data worked out a dynamic supply chain model to improve food tracking. Recently, Big Data is emerging as a crucial IT for instructing decisions in food supply chain. In order to differentiate and identify final food products, Ahearn et al. (2016) simulated environmental sustainability and food safety to improve food supply chain by using the consumer demands big data. This paper features a sustainability metric in agricultural production.
For practical data-driven system, various data are captured and collected to decision-makings in FSCM. Papathanasiou and Kenward (2014) produced a top level environmental decision support system by using the data collected from European food supply chain. It is found that socio-economic aspects have more influences on effective environmental decision support than technical aspects. Martins et al. (2008) introduced a shelf-life dating complex systems using sensor data to monitor, diagnose and control food quality. As the increasing focus on healthy diet, food composition and dietary assessment systems are significant for nutrition professionals. Therefore, Pennington et al. (2007) developed a system using the appropriateness of data for the intended audience. Most food and nutrition professionals will be beneficial from educating themselves about the database system. Perrot et al.(2011) presented an analysis of the complex food systems which are using various data such as supply chain dynamics, knowledge, and real-time information to make different decisions in FSCM. Tatonetti et al. (2012)illustrated a data-driven prediction system which is used for drug effects and interactions that US Food and Drug Administration has put great effects on improving the detection and prediction. Ahn et al. (2011), given increasing availability of information from food preparation, studied a data-driven system for flavor network and food pairing principles. Jacxsens et al. (2010) using actual microbiological food safety performance data designed a food safety management system to systematically detect food quality. The diagnosis is achieved in quantitative to get insight in the food businesses in nine European companies. Karaman et al. (2012) presented a food safety system by full using of data from plants where white cheese, fermented milk products and butter are produced. A case study from a Turkish dairy industry is demonstrated the feasibility and practicality of the presented system. In order to assess the lifecycle for sewage sludge and food waste, a system based on anaerobic codigestion of the organic fraction of municipal solid waste and dewatered sewage sludge was introduced (Righi et al., 2013). Environmental performances of various scenarios in the NE Italy case studies are evaluated to show energy saving using the data-driven system. Jacxsens et al. (2011) introduced a sort of tools for the performance examination and improvement of food safety management system by the support of food business data. These tools are able to help various end-users to selection process, to improve food safety, and to enhance performance. Food safety management systems usually use traceability and status data to examine food quality and freshness. Tomašević et al. (2013) took the Serbian meat industry for example to report food safety management systems implementation from 77 producers. Laux and Hurburgh (2012) reported a quality management system using food traceability data like maintain records for the grain scrutiny. A traceability index is used to quantify a lot size of grain in an elevator in this paper. Herrero et al.(2010) introduced a revisiting mixed crop-livestock system using farms’ data to achieve a smart investment in sustainable food production. By carefully consider the inputs of fertilizer, water, and feed, waste and environmental impacts are minimized to support farmers to intensify production. Tzamalis et al. (2016) presented a food safety and quality management system used in 75 SME by using the production data from the fresh-cut producing sector. This paper provides a best practice score for the assessment to ensure food quality and safety.
This section summarizes current challenges and highlights future perspectives in supply chain network structure, data collection, decision-making models, and implementations.
5.1Supply chain network structure
Food quality and safety heavily rely on an efficient and effective supply chain network structure. As the increasing globalization demands for more healthy and nutritious food, current structure is facing several challenges. First, the concentration of design and development of a food supply chain network structure is placed upon a sole distribution system or a WMS. Mixed-integer linear programming models are widely used to suggest proper locations and distribution network configurations (Manzini and Accorsi, 2013). An entire and global structure is necessary. Second, optimizations are always considered within a network structure. However, the common considerations are planning, scheduling, profit and cost. Environmental impacts and sustainable performance are omitted. As increasing consumptions of various resources, a sustainable supply chain network structure considering waste reduction and greenhouse gas emissions is needed. Third, with the development of advanced technologies such as IoT, traditional network structure is no longer suitable for facilitating the food supply chain operations because large number of digital devices, sensors, and robots are equipped along the supply chain. Thus, an innovative and open structure for FSCM is required.
Future structure for food supply chain network will be focused on the following directions so as to address current challenges and meet future requirements:
- An integrated global architecture: the final goal of this architecture is to control global food chain in both optimal and interdependent levels to make involved stakeholders for a closed-loop management and scrutiny. For achieving this purpose, new conceptual frameworks, effective supporting tools, integrated models, and enabled technologies are needed further investigation (MacCarthy et al., 2016; Talaei et al., 2016).
- Sustainable food supply chain: in the future, sustainable business in food industry can be harvested by reducing the environmental impacts, enhancing food waste recycling, and strengthening facilities sharing. New mechanisms and coordinated development along with other industries like manufacturing and economy are basic supports for achieving the sustainability (Green et al., 2012; Irani and Sharif, 2016; Lan and Zhong, 2016).
- Physical internet (PI) for FSCM: PI is an open global logistics system by using encapsulation, interfaces, and protocols to convert physical objects into digital items to achieve operational interconnectivity (Montreuil, 2011). Using the PI principle, FSCM for food handling, movement, storage, and delivery could be transformed toward global logistics efficiency and sustainability.
5.2 Implementations
FSCM implementations from real-life industries are based on cutting-edge technologies which are used for addressing some issues faced by food supply chain. Reported cases from literature mainly concentrated on verifying some hypothesis and presenting the improvements after using an IT system (Canavari et al., 2010; Soto-Silva et al., 2016). Few studies highlighted the natural characteristics of food supply chain or generic issues summarized from a set of companies so that the essence of FSCM could be figured out. After that, suitable technologies can be picked up to work out the solutions for the company or involved parties in food supply chain. Regarding the complexity of food supply chain, some important issues involving waste, re-use of resources, facility sharing, greenhouse gas emissions, and holistic lifecycle management are still unaddressed (Genovese et al., 2017). Take food waste for example, about 40 percent of total food produced in the USA goes as waste yearly which is equivalent of $165 billion (Pandey et al., 2016). Such vast wasted food not only physically influences our environment by polluting the water, but also significantly increases the CO2 emission since large number of pollution will be generated when they are deteriorating. Thus, reduction of food waste requires the actions at different echelons within food supply chain like food production, delivery, storage, retailing, and recycling. Regarding different echelons, associated solutions such as food production management system, WMS, logistics management system, etc. should be highly integrated in terms of data sharing and seamless synchronization.
Emerging cutting-edge techniques may contribute to system integration in the near future. First, Cloud technology has been used to integrate segregated sector using minimum resources. It allows involved stakeholders to access various services via software as a service, platform as a service, and infrastructure as a service (Singh et al., 2015). Through Cloud-enabled solution, the information sharing and collaborative working principle could be achieved by using basic computing and internet equipment. Second, IoT technologies like Auto-ID and smart sensors have been widely implemented in manufacturing and aerospace industry (Zhong, Li, Pang, Pan, Qu and Huang, 2013; Whitmore et al., 2015). IoT-based solutions for FSCM are able to provide an entire product lifecycle management via real-time data capturing, logistics visibility, and quality traceability. Additionally, within an IoT-based environment, every objects with sensing, networking and calculating ability can detect and interact with each other to facilitate logistics operations and decision-making in a fashion that is ubiquitous, real-time, and intelligent. Third, Big Data Analytics for FSCM has received increasing attention since it is able to deal with immense data generated from food supply chain. Big Data Analytics can help food companies to make graphical decisions with more accurate data input by excavating hidden and invaluable information or knowledge which could be used for their daily operations. With such information, ultimate sustainable food supply chain could be realized by optimal decisions.
In the future implementation, giant companies play important roles in leading the food supply chain toward a green and sustainable direction. To this end, collaborations with green relationships could lead to a win-win situation that large companies will get the economic benefits, and in turn the food supply chain members like SMEs could also be benefited. That green relationship is based on the joint value creation by using new business models in terms of internal and external green integration which will be enabled by advanced technologies (Chiou et al., 2011; Gunasekaran et al., 2015). So these companies may take initial actions to be equipped by advanced IT systems, while up-stream and down-stream parties within food supply chain can follow up for a green future.
Finally, the implementations need the involvement of government bodies which are going to work out strategic plans for guiding and supporting various enterprises toward a better future. Thus, Big Data Analytics is extremely important for these bodies to figure out up-to-data statistics report, current status of a food supply chain, and industrial feedbacks. Further to identify the strategies, they can use advanced prediction models or data-driven decision-making systems for assisting deeper analysis. As a result, each individual end-user could be beneficial from future implementation.
As the increasing awareness of food quality, safety, and freshness, FSCM is facing ever pressure to meet these requirements. How to upgrade and transform current FSCM to suit the ever increasing demands in the future? This paper presents a state-of-the-art review in FSCM from systems, implementations, and worldwide movements. Current challenges and future perspectives from supply chain network structure, data collection, decision-making models, and implementations are highlighted.
Based on the reviewed papers, some ideas and observations are significant for academia and industrial practitioners:
- advanced technologies like Big Data Analytics, Cloud Computing, and IoT will be employed to transforming and upgrading FSCM to a smart future;
- data-driven decision-makings for FSCM would be adopted for achieving more sustainable and adaptive food supply chain; and
- FSCM implementations will be facilitated by the cutting-edge technologies-enabled solutions with more user friendliness and customization.
