Report on findings on transportation and logistics of selected food value chains - Salmon to fillet case study

• Transportation has significant impact on food costs and the environment. It is a major contributor to carbon emissions, accounting for almost a quarter of the CO2 emissions in the EU, of which 30% is attributed to the food sector.

• This deliverable addresses the modelling of food chains’ transportation and logistics. It develops a robust model for policy support, which is applied to a specific case as a worked example. The approach can be used to model the transport and logistics of other food supply chains, given data availability.

• The mathematical modelling aims to optimise the cost and effectiveness of logistics operations. It also allows for the integration and consideration of environmental aspects within transportation, processing and distribution operations.

• Specifically, the deliverable focuses on the development of a logistics mathematical model using Atlantic salmon as an exemplary example of a globally integrated food supply chain. A Norwegian salmon exporter was engaged to supply data for validating the mathematical model.

• The model follows a multi-objective optimization approach that captures the trade-off between total logistics cost and the environment. It has two objectives. Firstly, to minimize total costs associated with transportation, fuel consumption, inventory holding, processing and residuals/waste. Secondly, to reduce CO2 emissions incurred by production at plants, transportation from suppliers to plants, and transportation from plants to customers.

• Constraints related to supply, processing capacity, storage capacity, demand, carbon emissions, inventory balancing, transportation capacity, and different modes of transportation between different types of plants and facilities are also consider within the model.

• Model development, validation and policy recommendation occurred in four stages: (i) mapping supply chain linkages and product flows, (ii) designing the mathematical model, (iii) data collection for parameters of the model and (iv) model validation and deriving policy recommendation.

• Before modeling, consultation with salmon supply chain actors occurred as a first step to map the supply chain linkages. This involved expert interviews with VALUMICS partners.

• Based on the mapping of the supply chain, a mathematical model was developed. However, given the complexity of the supply chain and the limited information that can be drawn from a single company which completely covers both the supply and the demand ends of the value chains, the model was divided into two stages (Model N1 and N2)

• First it optimises the supply chain network from salmon farms, abattoirs, primary processing plants, secondary processing plants and wholesalers so to meet the demand of the Secondary Processing Plants and Wholesalers for Fresh HOG (Head-on-Gutted) product (Model N1) (farm to wholesaler).

• Second, it addresses the supply chain from the secondary processing plants and wholesalers to retailers. The secondary processing plants process HOG into whole fillet, salmon by-products and some residual amount so to meet the demand of retailers (Model N2) (wholesaler to retailer).

• An additional model (Model M) allows for the optimisation of the overall supply chain network where, for example, a Company X tries to meet the demand of retailers in different time periods (farm to retailer).

• A transportation scenario analysis was also conducted by considering options for various maritime transportation routes from primary processing plant to secondary processing and primary processing plant to various wholesalers.

• The results from the three models highlight that it is essential for any company to optimise the overall supply chain network system (from salmon farms to retailers), as the total cost for model M is relatively much lower than the combined total cost of N1 and N2.

• Each model also shows that the supply chain network is sensitive to fuel cost and consequently, fuel consumption and distances between actors across the supply chain.

• Environmental impact is generally measured by fuel consumption during operations and in the case of food chain, transportation and distribution are important contributors via the use of fuel-based vehicles, sea vessels and/or airplanes.

• The scenarios analysis highlights the importance of adopting maritime transportation routes in terms of significantly reducing the total cost, fuel cost and overall carbon emission. Hence shifting certain logistics operations from road to maritime transportation from the perspective of economic and environmental benefits are advocated.

• For short to medium distances (vans, trucks, rails and sea vessels) that covers transportation trips to reach airport hubs and big cities, lowering CO2 emissions depends on the emissions ratio (the relative emissions impact of delivery vehicle when compared to personal vehicle – mostly applied in urban logistics) and customer density.

• For long distance transport (air), environmental improvement can be mainly achieved through technological development and this has been well supported by research dedicated specifically to address EU aviation industry challenges.

• The models are developed for a planning horizon consisting of discrete time periods, aiding the possibility of studying demand and supply uncertainty and its consequences in supply chain decision making. Hence, they help decision makers to identify the changes in a supply chain network when different transportation routes are adopted (for example whether maritime routes can be adopted or not in place of road/rail transportation, to address environmental concerns related to fuel consumption and carbon emissions).

• The models are valuable for policy makers in terms of understanding the costs and emissions associated with different food supply chains, as well as the effects of particular policy interventions and market developments (e.g. variation in demand, fuel costs, emission and waste constraints).

• They can aid supply chain managers to make decisions regarding the amountof inventory to be kept in different time periods.