Search results

This paper provides practical advice for archaeologists and heritage specialists wishing to use ML approaches to identify archaeological features in high-resolution satellite imagery (or other remotely sensed data sources). We seek to balance the disproportionately optimistic literature related to the application of ML to archaeological prospection through a discussion of limitations, challenges and other difficulties. We further seek to raise awareness among researchers of the time, effort, expertise and resources necessary to implement ML successfully, so that they can make an informed choice between ML and manual inspection approaches.

Automated object detection has been the holy grail of archaeological remote sensing for the last two decades. Machine learning (ML) models have proven able to detect uniform features across a consistent background, but more variegated imagery remains a challenge. We set out to detect burial mounds in satellite imagery from a diverse landscape in Central Bulgaria using a pre-trained Convolutional Neural Network (CNN) plus additional but low-touch training to improve performance. Training was accomplished using MOUND/NOT MOUND cutouts, and the model assessed arbitrary tiles of the same size from the image. Results were assessed using field data.

Validation of results against field data showed that self-reported success rates were misleadingly high, and that the model was misidentifying most features. Setting an identification threshold at 60% probability, and noting that we used an approach where the CNN assessed tiles of a fixed size, tile-based false negative rates were 95–96%, false positive rates were 87–95% of tagged tiles, while true positives were only 5–13%. Counterintuitively, the model provided with training data selected for highly visible mounds (rather than all mounds) performed worse. Development of the model, meanwhile, required approximately 135 person-hours of work.

Our attempt to deploy a pre-trained CNN demonstrates the limitations of this approach when it is used to detect varied features of different sizes within a heterogeneous landscape that contains confounding natural and modern features, such as roads, forests and field boundaries. The model has detected incidental features rather than the mounds themselves, making external validation with field data an essential part of CNN workflows. Correcting the model would require refining the training data as well as adopting different approaches to model choice and execution, raising the computational requirements beyond the level of most cultural heritage practitioners.

Improving the pre-trained model’s performance would require considerable time and resources, on top of the time already invested. The degree of manual intervention required – particularly around the subsetting and annotation of training data – is so significant that it raises the question of whether it would be more efficient to identify all of the mounds manually, either through brute-force inspection by experts or by crowdsourcing the analysis to trained – or even untrained – volunteers. Researchers and heritage specialists seeking efficient methods for extracting features from remotely sensed data should weigh the costs and benefits of ML versus manual approaches carefully.

Our literature review indicates that use of artificial intelligence (AI) and ML approaches to archaeological prospection have grown exponentially in the past decade, approaching adoption levels associated with “crossing the chasm” from innovators and early adopters to the majority of researchers. The literature itself, however, is overwhelmingly positive, reflecting some combination of publication bias and a rhetoric of unconditional success. This paper presents the failure of a good-faith attempt to utilise these approaches as a counterbalance and cautionary tale to potential adopters of the technology. Early-majority adopters may find ML difficult to implement effectively in real-life scenarios.

Unlike many high-profile reports from well-funded projects, our paper represents a serious but modestly resourced attempt to apply an ML approach to archaeological remote sensing, using techniques like transfer learning that are promoted as solutions to time and cost problems associated with, e.g. annotating and manipulating training data. While the majority of articles uncritically promote ML, or only discuss how challenges were overcome, our paper investigates how – despite reasonable self-reported scores – the model failed to locate the target features when compared to field data. We also present time, expertise and resourcing requirements, a rarity in ML-for-archaeology publications.

After completion of the case study, the students will be able to identify and discuss competition using Porter’s five forces, analyse and understand the enablers and challenges that impacted Rocky Brands’ growth and recommend a solution in relation to Rocky Brands’ growth strategy.

This case study investigates Rocky Brands, a South African manufacturer and distributor of cleaning products in the retail market. The case was set in November 2022 and highlights the important events ranging from the company’s founding in 2011 up until 2022. This case aims to study strategy in the South African fast moving consumer goods industry. At the time of writing the case study, Rocky Brands was operating across South Africa, with their main manufacturing warehouse in Johannesburg and a subsidiary manufacturing warehouse in Durban. They were changing the Durban warehouse to a distribution warehouse, as they planned to manufacture primarily from a bigger warehouse in Johannesburg. Rishav Juglall, the main protagonist, is the founder and managing director of Rocky Brands. Rocky Brands imports and redistributes several of the brands that the company sells, including Weiman’s, Wright’s and Goo Gone. They also manufacture their own line of products in South Africa under the Oakmont brand. Juglall acknowledges that their sales and revenue have grown yearly, but they have recently saturated the market and reached a plateau. Juglall needs to determine whether he should diversify into Africa, expand his product range or enter the market for private label cleaning products.

The case study’s primary focus is on strategy in an emerging market. This case study is suited to undergraduate students studying Porter’s five competitive forces, SWOT analysis (see teaching note exhibit) or the Ansoff matrix in the fields of strategy, marketing or macroeconomics. This case study can be taught in courses such as decision-making, environment of business, leadership or strategic implementation. The case study will teach students how to apply the frameworks to a business and assist students in determining which option is best for the business.

Teaching notes are available for educators only.

CSS 3: Entrepreneurship.

This study aims to deepen the understanding of the challenges and implications entailed by deploying mobile clinics in conflict zones to reach populations affected by violence and cut off from health-care services.

This research combines an integrated literature review and an instrumental case study. The literature review comprises two targeted reviews to provide insights: one on conflict zones and one on mobile clinics. The case study describes the process and challenges faced throughout a mobile clinic deployment during and after the Iraq War. The data was gathered using mixed methods over a two-year period (2017–2018).

Armed conflicts directly impact the populations’ health and access to health care. Mobile clinic deployments are often used and recommended to provide health-care access to vulnerable populations cut off from health-care services. However, there is a dearth of peer-reviewed literature documenting decision support tools for mobile clinic deployments.

This study highlights the gaps in the literature and provides direction for future research to support the development of valuable insights and decision support tools for practitioners.

This introductory chapter traces some of the impact COVID-19 has had on education in different global contexts. It traces the history of the Universitas 21 Schools of Education Deans Group and its Forum for International Networking in Education and sets the context for the subsequent contributions in this book. It provides points of reflection for scholars and institutions as they traverse through the ongoing challenges of the pandemic.

We study five exogenous shocks: climate, violence, price hikes, spoilage and the COVID-19 lockdown. We analyze the association between these shocks and trader characteristics, reflecting trader vulnerability.

Using primary survey data on 1,100 Nigerian maize traders for 2021 (controlling for shocks in 2017), we use probit models to estimate the probabilities of experiencing climate, violence, disease and cost shocks associated with trader characteristics (gender, size and region) and to estimate the probability of vulnerability (experiencing severe impacts).

Traders are prone to experiencing more than one shock, which increases the intensity of the shocks. Price shocks are often accompanied by violence, climate and COVID-19 shocks. The poorer northern region is disproportionately affected by shocks. Northern traders experience more price shocks while Southern traders are more affected by violence shocks given their dependence on long supply chains from the north for their maize. Female traders are more likely to experience violent events than men who tend to be more exposed to climate shocks.

The data only permit analysis of the general degree of impact of a shock rather than quantifying lost income.

This paper is the first to analyze the incidence of multiple shocks on grain traders and the unequal distribution of negative impacts. It is the first such in Africa based on a large sample of grain traders from a primary survey.

In this research, we explore the adverse impact of foreign ownership on operational security, a critical operational implication of the liability of foreignness (LOF).

The empirical analysis is based on a multi-country dataset from the World Bank Enterprises Survey, which contains detailed firm-level information from over 8,902 firms in 82 emerging market countries. We perform a series of robustness checks to further confirm our findings.

We find that a high ratio of foreign ownership is associated with an increased likelihood of security breaches and higher security costs. Our results also indicate that high levels of host countries’ institutional quality and firms’ local embeddedness can mitigate such vulnerability in operational security.

This study is one of the first to uncover the critical operational implication of the LOF, indicating that a high ratio of foreign ownership exposes firms to operational security challenges.

What kinds of support do interstate rivals provide to domestic actors in ongoing civil wars? And how do domestic actors utilize the support they receive? This chapter answers these questions by comparing Iranian and Saudi military and non-military (mediation, foreign aid and religious soft-power promotion) support to the Houthis and to the Government of Yemen (GoY) during the Saada wars (2004–2010) and the internationalized civil war (2015–2018). It also focuses on the processes through which the GoY and the Houthis have utilized this support for their own strategic purposes. This chapter applies a structured, focused comparison methodology and relies on data from a review of both primary and secondary sources complemented by 14 interviews. This chapter finds that there were less external interventions in the conflict in Saada than in the internationalized civil war. During the latter, a broader set of intervention strategies enabled further instrumentalization by domestic actors, which in turn contributed to the protracted nature of the conflict. This chapter contributes to the literature on interstate rivalry and third-party intervention. The framework of analysis is applicable to civil wars that experience intervention by rivals, such as Syria or Libya.