A hierarchical data architecture for sustainable food supply chain management and planning

Highlights

• Is developed within the European Union and Mediterranean Project FUTUREMED.

• Provides a framework to develop data-driven analyses for sustainable food operations planning and control.

• Addresses micro/macro aspects of the Food Supply Chain through a hierarchical data architecture and a tailored platform.

• Supports the design and planning of sustainable food distribution activities.

• Exemplifies an application of an Italian fruit trader considering opening a warehouse to serve new markets.

Abstract

The agro-food industry is one of the largest parts of the European Union's economy and faces economic and environmental stresses. While food traceability systems (FTSs) inform supply chain actors of product and logistical attributes, large scale implementations are scarce and are do not support active decision making. We present a framework developed for FUTUREMED project used to perform a data-driven analysis that considers both micro and macro aspects of a food supply chain (FSC). With its comprehensive multiple-depth data architecture incorporated within a tailored decision-support platform, this framework and the resulting decision-support tool is the first to move beyond simple traceability implementation to the sustainable planning of food logistics, bridging the gap between research techniques and real-world data availability. We define KPIs that measure a subset of economic and environmental factors to quantify the impact of logistical decisions. We validate the framework with the case study of an Italian fruit trader that is considering opening a new warehouse. We conclude by suggesting that this framework be applied to more complex case studies and be enhanced through including more dimensions of sustainability.

Introduction

The agro-food industry is a leading component of the EU economy, and governments are increasingly concerned about the long term sustainability of this sector (Notarnicola et al., 2017a, b). Others have addressed stressors of food system sustainability as follows. The exploitation of natural resources like water and soil (Kummu et al., 2012), resultant GHGs emissions (Sala et al., 2017), competition for land between food and biofuels (Rathmann et al., 2010; Fischer et al., 2010a,b; Cobuloglu and Büyüktahtakın, 2015), use of genetically modified organisms (GMOs), pesticides and chemical fertilizers (Gerdes et al., 2012; Roy et al., 2009; Boye and Arcand, 2013; McLaughlin and Kinzelback, 2015) are just few of the environmental issues affecting the food ecosystems. Food sustainability has a social as well as environmental dimension: volatile prices, bottlenecks, and food waste, losses, or contamination influence negatively impact actors and stakeholders' economic wellbeing (Garrone et al., 2014; Lebersorger and Schneider, 2014; Aiello et al., 2014). Furthermore, food insecurity contributes to social problems such as hunger and obesity (Fan and Brzeska, 2016; Thomas, 2010; Sadler et al., 2016).

Although these issues independently affect the environmental, economic, and social dimensions of sustainability, they are each exacerbated by ignoring the role operations play throughout the food supply chain (FSC) and the resultant interdependencies and externalities. Better information is crucial for improving food sector sustainability, and others (Hu et al., 2013; Etemadnia et al., 2015; Gwanpua et al., 2015) have demonstrated how increased information can drive such improvements. Higher visibility and transparency of food operations would enable precise performance assessments, resulting in more accurate planning and control of the production and distribution activities to the benefit of all stakeholders (Aiello et al., 2015; Accorsi et al., 2018).

Food traceability systems (FTSs) inform consumers and other supply chain actors of product attributes, shelf-life, and logistics aspects (e.g., delivery time) (Dabbene and Gay, 2011). Although FTSs could enhance knowledge about the FSC, barriers limit their diffusion (Bosona and Gebresenbet, 2013). Implementing a traceability system is expensive and requires expertise beyond most FSC companies (Bottani and Rizzi, 2008). The food sector is comprised of thousands of small actors that are neither horizontally nor vertically integrated. Lastly, information uncertainty couples with a lack of standardized data acquisition and communication protocols. Thus, effective implementations of FTSs are rare.

Encouraging the diffusion of FTSs to shed light on FSC ecosystems requires specifying which information to track (Tsolakis et al., 2014). This paper illustrates such a framework for FSC data collection that builds a structured knowledge on food processes and supports integrated, data-driven planning. The underlying work is from the European Union and Mediterranean Project FUTUREMED (MED/2007–2013–FutureMED) under the grant of the European Regional Development Fund. Their goal is mapping and optimizing agro-food supply chains for generic food products (i.e. both raw and processed) in the Mediterranean region through adopting Information and Communications Technology (ICT) in the agro-food sector, connecting supply chains actors. Note the proposed framework is intended for generic food products but remain open to others. Indeed, the characteristic that makes food supply chains distinctive and challenging even more than perishability is the extreme fragmentation of the sector, which counts hundreds of thousands of small producers who supply tens of thousands of intermediate players, who, in turn, serve thousands of selling points and hundreds of retailers, each owning a distribution chain. Furthermore, this hierarchy couples with that of decision-makers and of macro-micro issues which reflects the strategic, tactical and operational nature of supply chain planning. In such highly spread and fragmented environment, typical approaches to FTS are infeasible without first providing a comprehensive and integrated data architecture to cope with different processes to track and decision levers to handle. This assumption motivated our research.

The remainder of this paper is organized as follows. Section 2 summarizes a literature review. Section 3 defines the framework and the associated database for FSC data collection, storage, and manipulation. Section 4 illustrates how the framework enables data-driven analysis and supports decision-making for a case study. Section 5 concludes the paper by discussing the obtained results and considers potential further applications for the framework.