The COVID-19 pandemic brought about disruptive consequences to many people’s livelihoods around the world. The package of restrictions to contrast the health crisis caused a contraction of income and employment, on the demand side, and a disruption of domestic and global value chains, on the supply side. In many low-income economies, the crisis exacerbated an already fragile situation, raising concerns in terms of food insecurity and malnutrition. However, given the peculiar characteristics of the COVID-19 shock, not all individuals are expected to be affected in the same way. Nevertheless, evidence of the ultimate impact on food security, and the mechanisms of transmission, is still scarce. This thesis aims to address this literature gap, providing evidence for three African countries. Specifically, the study analyses the change in terms of food production and food consumption, as well as their relationship, in the aftermath of the COVID-19 outbreak, disentangling the heterogeneous impact over different types of households and different segments of the food value chain. To answer the proposed research questions, the study uses the most appropriate econometric techniques, which include a longitudinal model with household fixed effects, a structural equation model, and a cross-sectional model. What emerges is that the COVID-19 crisis severely impacted both household employment and income in 2020, the more so the longer the time length from the pandemic onset. The shock operated through two main channels of transmission, namely food value chain disruption and job loss, ultimately affecting household food security and child nutrition. The study also highlights the importance of considering the specific context under analysis and distinguishing between different types of households, specifically their market positioning when considering agricultural households.

Food processing micro- and small-scale enterprises (MSEs) play an important role in the national economic development of Tanzania. Though many of them have great growth potential, they face a number of constraints hindering further development, and large amounts of cooking oil are imported each year. The aim of thesis has therefore been to identify and analyse the different factors affecting these MSEs in order to find out which the major growth challenges are. The case study is mainly based on individual semi-structured interviews with sunflower oil processors and farmers in Babati districts, conducted in February and March 2016, and earlier research and studies on the topic of MSE growth make up the theoretical framework used for analysis of the data. The findings show that there are indeed numerous challenges facing these processors, and the major constraint was found to be lack of capital; an issue causing or worsening a majority of the other challenges at hand. Other problems are related to raw material, equipment & electricity for processing, regulations, market accessibility, and competition. These obstacles need to be overcome in order to enable the industry's expansion within and outside of Tanzania, and further research is recommended.

Purpose: Information management is crucial for a food manufacturing company as it increases productivity, lowers cost, and enables traceability as well as data-driven decision-making. A lack of information management leads to consequences such as lack of customer demand and requirements, which result in high inventory or stockout. The purpose of this study is to enable digitalized information management in food value chains, and this study aims to know which information needs to be considered for data-driven decision-making. The purpose and aim align with adapting to the current trends of industry 4.0 and digitalized information management. To fulfill the purpose, two research questions are formulated: (1) What type of information needs to be considered for data-driven decision-making in food value chains? and (2) How should the parts of the value chain be digitalized to enable data-driven decision-making in food value chains?  Method: The chosen approach for this study was an inductive approach and the chosen strategies were a literature review and a single case study in the Swedish food manufacturing industry. To gain an understanding of the case company’s value chain, interviews were used as a data collection technique. The interviews were then analyzed and then combined with the literature review to answer both the research questions. The quality of the study was evaluated by using thetrustworthiness criteria. There were also five principles, when it comes to research ethics, that were used during this study.   Findings: After the conducted interviews, the relevant actors were identified and mapped in the case company’s value chain. It was found that the company used a lot of manual information management procedures, which led to several challenges for data-driven decision-making within the company. The mapping revealed which information was relevant for the respective actors within the case company. This provided a starting point from which empirical and theoretical data were studied to address these challenges and answer the two research questions and fulfill the purpose of this study.   Analysis: To build a framework for digital information management, it was necessary to customize existing equipment and resources, adapt the process to the specific industry, and move toward the concept of industry 4.0. With these elements in mind, the framework was created using data acquired from both the literature research and the case study. The framework consists of a loop, that allows for continual improvement.   Conclusion: Applying the correct technology is important for digitalized information management, and the food manufacturing industry suffers from limited technologies in this aspect. This thesis informs the reader on how to digitalize information management. The academic contribution is a theoretical framework for the digitalization of information management using industry 4.0 concepts, which can support companies to generate revenue. / <p>Externt samarbete utesluts, eftersom företaget inte vill skylta med sitt namn. </p>

Artificial intelligence (AI) holds immense potential to revolutionize the global food system, driving sustainability, enhancing efficiency, and addressing food security challenges. However, the successful integration of AI in the food system demands a deep understanding of the complex interplay between technology, social factors, economic considerations, and ethical implications. This study explores the opportunities and challenges in implementing AI technologies within the Swedish food system, focusing on primary production. The research utilizes an expanded Socio-Technical System Theory (STST) framework, incorporating economic and ethical dimensions alongside the traditional social and technical levels. Literature review and semi-structured interviews provide insights into the dynamics of AI adoption in the Swedish context. The findings reveal that AI adoption in the Swedish food system is currently in the early adopter phase, with broad range applications. However, the study also uncovers significant barriers to widespread AI adoption, including the lack of suitable business models, fragmented data sharing infrastructures, and ethical concerns surrounding data privacy and ownership.  The analysis emphasizes the need for developing user-friendly interfaces, leveraging narrow AI applications, and establishing seamless data flow across the food value chain. The study contributes to the theoretical development of the Socio-Technical System Theory framework by demonstrating the importance of incorporating economic and ethical dimensions in understanding the complex dynamics of AI adoption in socio-technical systems. The findings also have practical implications for policymakers, industry actors, and researchers, emphasizing the significance of context-specific AI development, as well as the need for collaborative innovation processes. The research acknowledges its limitations, including the focus on primary production and the reliance on qualitative methods, and identifies potential areas for future research, such as comparative analyses across different food sectors and the use of quantitative approaches.  In conclusion, this study provides a timely and critical contribution to the understanding of AI's role and potential in transforming the Swedish food system. It indicates the need for developing suitable business plans, establishing data sharing platforms, and ensuring a harmonized data flow to harness the benefits of AI while navigating its challenges and risks, paving the way for a more sustainable and resilient food future. / Artificiell intelligens (AI) har enorm potential att revolutionera det globala livsmedelssystemet, driva hållbarhet, förbättra effektiviteten och hantera livsmedelssäkerhetsutmaningar. Men en framgångsrik integration av AI i livsmedelssystemet kräver en djup förståelse för det komplexa samspelet mellan teknik, sociala faktorer, ekonomiska överväganden och etiska implikationer. Denna studie utforskar möjligheterna och utmaningarna med att implementera AI-teknologier inom det svenska livsmedelssystemet, med fokus på primärproduktionen. Forskningen använder ett utökat ramverk för Socioteknisk Systemteori (STST), som inkluderar ekonomiska och etiska dimensioner tillsammans med de traditionella sociala och tekniska nivåerna. Litteraturöversikt och halvstrukturerade intervjuer ger insikter i dynamiken kring AI-antagande i svensk kontext. Resultaten visar att AI-antagandet i det svenska livsmedelssystemet för närvarande befinner sig i den tidiga antagandefasen, med breda tillämpningar. Studien avslöjar dock också betydande hinder för ett utbrett AI-antagande, inklusive bristen på lämpliga affärsmodeller, fragmenterade datadelingsinfrastrukturer och etiska bekymmer kring dataintegritet och ägande. Analysen betonar behovet av att utveckla användarvänliga gränssnitt, utnyttja smala AI-applikationer och etablera sömlösa dataflöden över hela livsmedelsvärdekedjan. Studien bidrar till den teoretiska utvecklingen av Socioteknisk Systemteori genom att visa vikten av att inkludera ekonomiska och etiska dimensioner för att förstå den komplexa dynamiken kring AI-antagande i sociotekniska system. Resultaten har också praktiska implikationer för beslutsfattare, branschaktörer och forskare, och betonar betydelsen av kontextspecifik AI-utveckling samt behovet av samarbetsinriktade innovationsprocesser. Forskningen erkänner sina begränsningar, inklusive fokus på primärproduktion och beroende av kvalitativa metoder, och identifierar potentiella områden för framtida forskning, såsom jämförande analyser över olika livsmedelssektorer och användningen av kvantitativa metoder. Sammanfattningsvis ger denna studie ett aktuellt och kritiskt bidrag till förståelsen av AI roll och potential i att transformera det svenska livsmedelssystemet. Den pekar på behovet av att utveckla lämpliga affärsplaner, etablera plattformar för datadelning och säkerställa ett harmoniserat dataflöde för att utnyttja AI fördelar samtidigt som man navigerar dess utmaningar och risker, och banar vägen för en mer hållbar och motståndskraftig livsmedelsframtid.