Food Network Analysis Project

Systems intrigue me for some reason.  Perhaps it is the mental challenge of understanding not only the elements involved, but also the interactions of those elements.  While going through a Nutrition Science certificate program, I became interested in the food system of the United States.  This interest was fueled by a 2015 publication, A Framework for Assessing Effects of the Food System, from the Institute of Medicine and National Research Council of the National Academies.  This publication stressed taking a systems-thinking approach to understanding the food system operating in the United States.  This combined with some coursework in network analysis using the software tool, R, prompted me to consider developing a network model of the food system.

Admittedly, I don’t have a specific question to answer with such a model.  Rather, I am more interested in visualizing the system to facilitate an understanding of how the elements interact and be better able to assess the challenges and impact of changing the system.

Systems

A system can be thought of as having three components: elements, interactions, and a purpose.  Elements are the agents or actors that can perform actions such as farmers, manufacturers, and retailers.  Interactions are the ways in which elements interact with each other.  And purpose is the reason for those elements to be acting and interacting, the function, if you will, of the system.  In the case of the food system, a purpose may be to provide enough nutritious food to support a given population in an ecologically sustainable way. The first paragraph of the Summary of the, A Framework for Assessing Effects of the Food System, states:

“The nation’s food system historically has seen remarkable success in providing the U.S. population with a varied, relatively inexpensive, and widely available supply of food.” 

Elements

A Framework for Assessing Effects of the Food System, provides a foundational representation of the actors in the food supply chain as well as the broader economic, biophysical, and sociopolitical contexts in which it operates.  Figures 2-1, 2-2, and 2-8 provide a starting point for defining the elements to be used in the model.  These figures also provide the initial conceptualization of interactions.

In R, an element is referred to as a vertex.  Other graph database software may refer to it as a node.  I prefer the term node.  The initial set of nodes (vertices) for my database are:

F             Food System elements (figure 2-1)

  FI          Farm input supply = capital and purchased inputs (e.g., stock, seeds, chemicals)

  FP         Farm production = production of whole plant and animal food products

  FH        First-line handlers = initial processing of whole foods such as washing, waxing, and milling

  FM       Food manufacturers = production of food products for consumers

  FW       Wholesale and logistics = storage, distribution, and transportation of food products

  FB         Food and beverage services = services providing prepared food products such as restaurants

  FS         Institutional buyers = providers of prepared food in institutional cafeteria setting

  FR         Retail food stores = provide food products for home preparation and consumption

  FN        Food banks = providers of food assistance to consumers

C             Consumer (figure 2-1)

  C           Consumer = the terminal step of the supply chain wherein food is purchased for consumption

N            Context = other systems that support or interact with Food System and Consumers (figure 2-8)

  NS        Social organizations = organizations that endeavor to change consumer and producer behavior

  NT        Science and technology = research and innovation in the food supply chain

  NB        Biophysical environment = the natural resource base for producing whole food products

  NP        Policies = local, state, and federal policies affecting the food system operations

  NM      Markets = marketing, lifestyle, cultural, health, and economic influences on consumer demand

The above initial set of nodes is labeled and structured in a hierarchy.  Such a structure accommodates successively deeper explorations, and representation of, component elements while maintaining alignment to higher order structure (e.g., Food System, Consumer, Context).

Interactions

Figure 2-1 provides a map of the flow of food through the supply chain; what can colloquially be termed, “from field to fork.”  The terminus of the chain, or flow of food, is the consumer.  This flow can be represented as lines connecting the nodes in the chain.  In R, such interactions are called an edge.  Edges not only link nodes, but they can also show a direction of flow which is appropriate for a supply chain representation.

Initial Network Visualization

Next Steps

1.      Develop links between supply chain nodes and context nodes, including potential feedback loops.

2.      Develop attributes of edges such as using the NOVA food processing classification system to capture how food is transformed through supply chain.

Reference:

IOM (Institute of Medicine) and NRC (National Research Council). 2015. A framework for assessing effects of the food system. Washington, DC: The National Academy Press.