Decision Tools and Management in the Fish Sector: Examining the psychological components of the Analytical Hierarchy Processes methodology and its effect on decision-making.

Europe’s fishing industry processes over five million tonnes of fish annually, of which, almost two-thirds ends up as side-streams resources. These side-streams are either used for low-value animal feed or disposed of, which is a costly endeavour for the companies involved. The impact of underutilized side-stream resources in the industry can be expressed in decreased high quality production of bioactive peptides for nutraceuticals, food, and feed applications as well as protein-based food ingredients, savoury ingredients, and mineral supplements.

This leaves the industry and society with inefficiency and a poorly utilized aquatic living resource biomass that could be used for higher value purposes such as securing affordable protein sources or developing bioactive peptides for nutraceutical products, pharmaceutical production, food, and feedapplications.

To meet these challenges, a main objective of current large-scale research projects, such as the EU Horizon2020 WaSeaBi project(https://www.waseabi.eu/), is to develop the state-of the-art in solving the barriers to the sound exploitation of aquatic living resources. This involves a focus on better utilization of fish side-streams through the development of storage solutions, sorting technologies and management tools to secure an efficient, sustainable supply system for fishery by-catches and side-stream utilization from aquaculture, fisheries, and the aquatic processing industries.

One such barrier is related to the fact that human decision-making, also in the food production industry, is inherently limited and more subjective rather than objective due to our bounded rationality when handling new and/or complex information. Therefore, there is a need for further development and utilization of robust and generic management tools to mitigate these flaws and support a more rational and unbiased decision-making process.

The purpose of this PhD thesis is therefore to illustrate how fish processing and seafood companies can optimize their decision-making processes by using decision tools to improve the sustainability and economic utilization of side-stream resources. This will mainly be achieved through application of an Analytical Hierarchical Process (AHP) support tool that can account for a decision-maker’s different Cognitive biases and minimize the influence that these have on the AHP tools results.

However, there is a current lack of a comprehensive method framework for support tools that focuses both on the identification of specific Cognitive biases, but also their correction or Debiasing in a AHP context. To this end, a novel method framework has been developed as part of the thesis called Debias Your Decisions (DYD) and implemented for testing within a fish processing company seeking to optimize their side stream utilization (paper 1 and 2).

The results of the testing of the novel DYD framework in both a theoretical simulation and real-world setting shows that the developed method was able to facilitate the identification of specific Cognitive biases. In addition, it provided a structured approach to the Debiasing process in a AHP context using already established semi-quantitative Debiasing techniques found within the Debiasing literature in a Multi-Criteria-Decision-Making (MCDM) context, of which AHP is a part of. These Debiasing techniques focuses on either correcting the weight elicitation scores or by directly intervening in the normalization index scores, both of which are central components for an AHP tools result generation.

By ensuring this structured approach through the DYD framework and thus implementation of relevant Debiasing techniques to correct the identified Cognitive biases, less negative impacts due to these Cognitive biases on the AHP tools results was achieved. These results also showed high robustness in the subsequent sensitivity analysis (paper 1 and 2).

The DYD framework was thus able to enable the decision-makers to engage in more rational and objective processes with an AHP tool, by providing a correction for their user ranking for the best-evaluated alternative within a side-stream scenario, to be in line with the experts ranking for the same scenario in terms of environmentally and economically efficient production. Thus, achieving a greater optimization of their side-streams (paper 1 and 2). This optimization is expected to benefit the companies both economically but also on environmental metrics in relation to compliance with different environmental regulations (paper 2).

This is a significant contribution to the Debiasing research area focused on AHP tools, since no prior framework has been developed that includes both the process of identifying the specific Cognitive or Motivational biases that can influence the AHP results, but also includes a structured approach on how to correct for these.

Important further improvements to the DYD framework can still be made such as:

• Expanding its application to other MCDM methods and tools.
• Improving the bias identification process towards an automated process, thus minimizing the conductor of analysis involvement.
• Expanding the current set of semi-quantitative Debiasing techniques to include more semi-quantitative techniques for specific biases such as: Planning bias, Affect heuristic, Sunk cost fallacy etc.

One of the main future perspectives from this thesis is the application of DYD to other industries that uses MCDM tools such as AHP, to test the frameworks generic application ability and to further test the frameworks robustness in other settings beyond the European fish processing and seafood industry (paper 2 and 3).

In addition, another future perspective is the need to develop a standardized set of guidelines on when and which Debiasing methods to use based on different conditional criteria in a given decision process (paper 3)