Business and Economics

Phd defences 2019

Interpersonal Trust in Vietnamese Companies: a study of organizational antecedents and effects - Dinh Thi Le Trinh (13/12/2019)

​Dinh Thi Le Trinh

  • 13 December 2019
  • Supervisors: Prof. Ann Jorissen and Prof. Koen Vandenbempt

Abstract

Developments in organizational sciences underscore the importance of interpersonal trust for sustaining individual and organizational effectiveness. As sub-characteristics of interpersonal trust, affective-based and cognitive-based trust in the supervisor form the basis of a desirable work climate between supervisors and subordinates; foster job attitudes and positive behaviors of subordinates, such as job satisfaction, organizational commitment, job performance and organizational citizenship behaviors, and increase an organization’s competitive advantage. Unfortunately, the study of the effects of affective-based and cognitive-based trust in the supervisor is still rather limited and needs further investigation, particularly in the literatures related to performance evaluation, role stress and leadership behavior. This PhD manuscript, therefore, investigates: (1) how supervisor’s performance evaluation practices affect subordinates’ affective-based and cognitive-based trust in the supervisor through role clarity; (2) how role stressors differently influence managerial employees’ affective organizational commitment through affective-based and cognitive-based trust in the supervisor and trust in the organization; and (3) how the supervisor’s leadership behavior impacts on the subordinates’ organizational commitment through their affective-based and cognitive-based trust in the supervisor. A survey of employees with managerial functions in Vietnamese companies allows us to test hypotheses related to the above questions. The research findings are as follows.

Regarding performance evaluation, when the supervisor practices participative, formal, diverse and controllable performance evaluation, subordinates directly increase their affective-based and cognitive-based trust in the supervisor.  Subordinates also increase their affective-based and cognitive-based trust in the supervisor partly through role clarity.

Regarding role stress, when managerial subordinates experience role ambiguity, a reduction in affective-based and cognitive-based trust in the supervisor is observed. However, only cognitive-based trust in the supervisor decreases the level of the affective organizational commitment. The managerial subordinates also reduce trust in the organization, leading to a decrease in their affective organizational commitment. Meanwhile, when managerial subordinates experience role conflict, they will decrease their cognitive-based trust in the supervisor, leading to a decreased level in their affective organizational commitment.

Regarding leadership behavior, the supervisor’s initiating structure behavior has a stronger effect on subordinates’ cognitive-based than affective-based trust in their supervisor, while consideration behavior has a more significant effect on affective-based than cognitive-based trust. However, it is only the subordinates’ cognitive-based trust that positively mediates the relationships between the supervisor’s leadership initiating structure and consideration behavior and their affective and normative organizational commitment.

Understanding IT Governance: Conceptualization and Enabling IT Business Value - Tim Huygh (05/12/2019)

​Tim Huygh

  • 5 December 2019
  • Supervisor: Prof. Steven De Haes

Abstract

An increasing (operational and strategic) dependence on IT for organizations results in a growing need for effective IT governance. Effective IT governance effectuates appropriate control over an organization’s current and future IT use, as to enable the creation and protection of IT business value. However, the research domain of IT governance is somewhat hampered by a lack of theoretical underpinnings for IT governance, which makes it difficult to explain from a theoretical perspective how effective IT governance should be organized and why.

In response, this PhD thesis investigates the concept of IT governance through Beer's Viable System Model (VSM). In that context, the research that was conducted serves multiple purposes. First, an organizing logic for IT governance is articulated that is grounded in the VSM and informed by extant IT governance research. Second, extending the discussion beyond the conceptual level, two in-depth case studies are presented that demonstrate how the VSM can be used as a lens for describing and diagnosing IT governance in practice. Third and finally, four empirical IT governance-related research studies are presented that are inspired by the VSM-based organizing logic for IT governance, which demonstrates that this logic can provide structure and theoretical underpinnings to IT governance-related empirical research.

For academics, this thesis contributes to the theoretical discourse on IT governance. It answers the question why IT governance can continue to fulfil its general purpose of creating and protecting IT business value. Furthermore, it provides strong theoretical underpinnings for how to organize effective IT governance. For practitioners, this thesis shows how the VSM can be used as a lens to describe and diagnose IT governance in practice. Furthermore, it provides insights on the IT governance mechanisms and practices that may be used to instantiate the required functions of an IT governance arrangement.

Cooperation between seaports concerning hinterland transport - Joost Hintjens (22/11/2019)

Joost Hintjens

  • 22 November 2019
  • Supervisors: Prof. Eddy Van de Voorde and Prof. Thierry Vanelslander

Abstract

Seaports are, on the one hand, crucial links in the global supply network. As such, they are important sources of added value, employment and welfare. But on the other hand, they need big investments, often made with public money, and they are places where external costs are generated that are not appreciated by the communities of which they are part. Society demands from ports that in return for this ‘licence-to-operate’ the ports operate as efficiently as possible. More and more voices demand that through cooperation, this efficiency is improved, and the external costs diminished. Others believe that independence and competition is the best guarantee to make ports as efficient as possible. This thesis studies the possibilities cooperation between competing seaports can offer with a focus on port authorities and how they can cooperate in extending the hinterland.

The viewpoint is that of society, to allow the benefits of all stakeholders to be considered. A conceptual model, based on societal cost benefit analysis, is developed where the welfare effects of different cooperation strategies are analysed. More in detail, the conceptual model focusses on the social costs and benefits of combining hinterland road cargo flows of cooperating ports into a bundled transport mode, thus lowering direct and external costs and increasing the market share of the cooperating ports. This is further developed into a empiricalized cost model that combines the EU hinterland at NUTS2 level with the road cargo flows of the 104 core TEN-T ports, concluding with a tool that enables the calculation of the direct cost benefits, the effect on the value of time and the potential external cost savings of any cooperation between the 104 core ports.

The model is applied to three cooperation case scenarios. The first one consists of a bundling of the road cargo flows of the ports of the recently created North Sea Port towards the NUTS2 region of Düsseldorf. The second case concerns the bundling of the flows of the ports of Rotterdam and Antwerp towards the region surrounding Cracow. The last case calculates the effect of the modal shift facilitated by the cooperation of the four Polish ports towards their main hinterland region. From this possible cooperation strategies are suggested for the different port actors.

Empirical essays on the effects of the economic cycle on human capital, health and fertility - Sofia Maier (21/11/2019)

​Sofia Maier

  • 21 November 2019
  • Supervisors: Prof. Sunčica Vujić and Prof. Guido Erreygers

Abstract

The ongoing booms and busts which modern capitalist economies face can influence individual well-being and household decision-making in multiple ways. Economic choice models suggest that income and substitution effects involved in the demand for human capital, health and children make the effect of the economic cycle theoretically ambiguous. Therefore, whether counter- or pro-cyclical patterns dominate is an empirical question.

Unfortunately, after more than a hundred years of empirical inquiry into these topics, most of the evidence still refers to a limited set of rich western economies. This thesis provides empirical evidence on the effects of the economic cycle on human capital, health and fertility through four empirical essays, covering developed and developing countries, and combining national-, regional- and individual-level data. It also explores the mechanisms driving these patterns, the presence of asymmetries, heterogeneities and scarring effects.

Results suggest that households systematically adapt their demand for human capital, health and fertility in response to booms and busts, and that these patterns are heterogeneous across countries as well as by age, gender and socio-economic background. Human capital formation and health are counter-cyclical in high and some middle-income countries (paradoxically, they improve in recessions). By contrast, pro-cyclical patterns are observed in the poorest countries, mainly driven by children and women. Fertility rates are in general pro-cyclical, with an inverted U-shape pattern: middle-income countries show the strongest responses. These heterogeneities are in general consistent with the expected relative importance of opportunity cost (substitution effect) and ability-to-pay considerations (income effect) across these groups.

Policy implications are diverse. For low-income countries expanding safety nets –e.g. through the expansion of social security systems, access to credit and public services– could play a key role in alleviating the adverse effects of crises on human capital and health. This can also help smoothen the propagation of shocks. For middle- and high-income countries, where safety nets are stronger, governments face the challenge of preventing the negative side effects of booms on human capital and health through the design of labour policies and public services aimed at reconciling the expansion of jobs with investments in these non-monetary dimensions of well-being. By contrast, the pro-cyclicality of fertility suggests that the private costs of reproduction constitute an important burden to the household budget, and especially for women who interrupt their labour careers in periods of high opportunity costs. In this way, gender inequalities in labour markets tend to be reproduced, highlighting the importance of national care services and policies designed to reconcile work-life balance for both women and men.

Maritime supply chain innovation: costs, benefits and cost-effectiveness of ICT introduction - Valentin Cârlan (20/11/2019)

​Valentin Cârlan

  • 20 November 2019
  • Supervisors: Prof. Dr. Christa Sys and Prof. Dr. Thierry Vanelslander

Abstract

ICT innovation has become a key prerequisite for maritime supply chain (MarSC) stakeholders to remain competitive. In this context, new technological advancements from early EDI-based (electronic data interchange) solutions to contemporary blockchain-based tools are continuously changing the market.

The MarSC actors need to decide which ICT solution to implement. In this process, own financial benefits are key. Equally so, the broader effects on other stakeholders participating in the chain are relevant. However, there is no comprehensive framework available to identify the costs and benefits of ICT innovation in the MarSC reported in the literature, nor applications that calculate their cost-effectiveness.

This research identifies the cost-effectiveness of integration practices introduced by ICT innovation in the MarSC. To do so, it develops, validates and applies a comprehensive costs and benefits framework taking into consideration key characteristics of ICT innovation. A collection of 44 cases provides further empirical evidence with regard to the cost and benefit elements they generate. Subsequently, four detailed cost-effectiveness analyses are carried out focusing on integration.

The first analysis shows how an internal data integration practice is cost-effective for a road transport company and reduces external environmental effects at the same time.

The second one studies horizontal data-integration practices in the MarSC. Focusing on data integration amongst hinterland transport operators, the analysis proves that taking a multi-disciplinary approach in building ICT innovation provides a more cost-effective outcome.

The third case study focuses on vertical data integration in the MarSC. It discusses the cost-effectiveness of three strategies to implement a blockchain-based application in the MarSC that facilitates the official documents transfer. This last analysis provides evidence with regard to the increase in a community’s competitive advantage, when a mixed data integration tool is used. This tool is developed in the framework of a port community system.

Overall, this dissertation provides new relevant insights with regard to costs and benefits generated by ICT innovation in the MarSC, and is valuable for both academia and industry.

Fostering technological change through regulations conducive to responsible mining of ocean mineral resources - Kris Van Nijen (18/10/2019)

​Kris Van Nijen

  • 18 October 2019
  • Supervisor: Prof. Steven Van Passel

Abstract

Metal demand continues to increase due to rising global population, urbanisation, and to meet society’s clean energy goals. Polymetallic nodules are rock concretions (the size of tennis balls) consisting largely out of minerals. They are found in the Pacific Ocean at 4,500 meters water depth. It is estimated that these nodules contain more nickel, cobalt and manganese than all land-based reserves combined. By 2020, the UN International Seabed Authority (ISA) will deliver the world’s first set of exploitation regulations for polymetallic nodules found on the seabed of the area beyond national jurisdiction (“the Area”). The ISA, an intergovernmental body to manage all mineral-related activities in the Area, was established by the Law of the Sea Convention (LOSC) in 1994. The LOSC designates its resources as the common heritage of (hu)mankind and calls for the development of the resources (by attracting investment and technology), and the protection of the marine environment for the benefit of (hu)mankind. Under an exploitation contract, contractors receive the right to exploit polymetallic nodules in return for a payment, which must be fair for both the ISA and the contractor. The ISA must ensure these proceeds are equitably shared among all members taking into particular consideration the interests and needs of developing States.

This dissertation provides valuable insights into the interaction between regulatory development and the impact on a contractors’ decision to invest in new “biased” technological change (i.e., technological change that lowers the environmental impact relative to the status quo), thereby contributing to the awareness of the ISA. The author hereby focusses on the five-year multi-stakeholder participation process (2015-2019), that led to the draft exploitation code. After section one offers a general introduction, section two provides a synthesis of workshop proceedings, ISA technical papers and scholarly literature, offering more detail on the multi-stakeholder process that led to a payment system. In section three, a stochastic techno-economic assessment is delivered to model the impact of different payment rates on the economic performance of a polymetallic mining project. In section four, a risk assessment evaluates the financial, technical & institutional risks and provides insights into the minimum (private) financial return required to attract investment. Chapter five analyses different technology and environmental policy instruments that foster biased technological change. Finally, the author concludes with new insights on the impact of the proposed regulations on technological change required for responsible deep sea mining.

Brand Communication on Social Networking Sites - Freya De Keyzer (08/10/2019)

​Freya De Keyzer

  • 8 October 2019
  • Supervisors: Prof. Nathalie Dens and Prof. Patrick De Pelsmacker

Abstract

Globally people have embraced social media as important communication channels. These platforms have also caught advertisers’ attention who try to engage with (potential) customers through brand communication on these platforms. Brand communication can be defined as every brand-related communication spread by both firms and users. The main objective of this dissertation is to study consumers’ responses to brand communication on social networking sites (SNSs). This objective is explored from two different angles by examining consumer responses to both personalized advertising and word-of-mouth.

Social networking site users reveal a large amount of personal information (e.g., location, interests, demographics). That information can be used by advertisers to tailor their messages to specific targets. Our findings suggest that the use of all personalization elements under study are able to elicit perceived personalization. However, interests seem to be the most important. Second, we are able to show a strong positive effect on consumer responses through perceived relevance, entertainment, and intrusiveness. Moreover, we show that when the SNS is liked or when consumers have the perception the SNS protects their privacy, then the indirect effect through perceived relevance becomes stronger. Finally, we found a negative impact of personalization through perceived creepiness which only occurred for a health website, governmental website and online newspapers, but not for commercial website.

Next to (personalized) advertising, we studied word-of-mouth communication on SNSs (sWOM). It is generally accepted that a positive (resp. negative) sWOM message leads to positive (resp. negative) consumer responses. This effect is strengthened when the message uses a factual, compared to an emotional, tone of voice. Moreover, this effect also becomes stronger when evaluating an sWOM message about a more hedonic compared to a more utilitarian service. This message valence effect can also be strengthened by relational characteristics, both with the sWOM sender and the SNS on which the message was posted. However, when considering both the interpersonal relational characteristics and the person-to-site characteristics, only the interpersonal relational characteristics moderate the message valence effect.

Design of Modular Structures for Evolvable and Versatile Document Management Based on Normalized Systems Theory - Gilles Oorts (01/10/2019)

​Gilles Oorts

  • 1 October 2019
  • Supervisor: Prof. Herwig Mannaert

Abstract

Despite technological and operational business advances over the past decades, organizations are still required to draft and manage documents. These documents can take a plethora of forms, such as books, spreadsheets, slide decks, manuals, legal contracts, emails, reports, etcetera. Although a lot of these documents have taken an electronic form, their structure is in essence still the same as their analogue and physical predecessors. Invoices are often just printed and sent by mail, after which they are opened and scanned by the receiving organization. Or instead of printing and handing out new operational procedures, they are often just exported as a pdf-file and saved on a server.

Despite the endless opportunities the revolution in Information Technologies (IT) has brought along, most efforts in document management were limited to just digitizing documents, i.e., transforming them from analogue to digital form as monolithic blocks. In addition, document management is faced with challenges such as content customization, the challenge to convert documents into value-adding products or services, repurposing internal documents for use on the internet and the management of the mountain of documents that are created daily.

In this dissertation, we show how this view of static documents that are a mere representation of their analogue predecessors is out-of-date. Instead, we present a view of multidimensional and ever-changing documents, based on the insights from modularity and Normalized Systems reasoning. An essential concept of our document management approach is that one has to consider the underlying artifacts that are described in documents. The structure of these artifacts is unconditionally connected to the structure of the documents they are described in. For documents to be modular and evolvable, one should therefore always first analyze its underlying artifacts. Ideally, the effort of modularizing documents in an effort to make them evolvable should be preceded by going through the same process for the described artifacts (although this is not a requirement).

The practical implications of this approach are discussed based on a case study of a document management system for study program documentation. To demonstrate the feasibility of our document management approach, a proof of concept application was developed.

What role for rail in urban freight distribution? - Katrien De Langhe (24/09/2019)

Katrien De Langhe

  • 24 September 2019
  • Supervisors: Prof. Dr. Eddy Van de Voorde and Prof. Dr. Christa Sys

Abstract

Many national and international institutions encourage the use of environment-friendly transport modes. Subsequently, local authorities take increasing measures to prevent negative transport-related externalities in urban areas. Hence, logistics service providers consider alternative ways to deliver goods in urban areas. Which alternative mode is appropriate depends on multiple factors, including the available transport infrastructure, the freight volume, the time of the transport, the measures taken by the authorities and the presence of congestion.

This doctoral thesis focuses on urban freight distribution by rail and the conditions for a successful implementation. The potential success is studied from a financial, economic and a socio-economic perspective for a dedicated freight vehicle, a freight wagon attached to a passenger vehicle and the transport of freight alongside passengers. The government’s viewpoint is adopted. A generic tool is created, by developing a tailor-made social cost-benefit framework and investigating the rail freight product in depth. The generic framework is applied to a case study for the use of a tram in the city of Antwerp.

The main lessons to be learned are twofold. Firstly, the good environmental performance of rail and the presence of congestion on the road network favours the shift from road to rail transport. The interference with passenger traffic, resistance from different stakeholders, initial investment needed and commitment of different stakeholders are disadvantageous for rail transport. Secondly, attaching a freight wagon to a passenger vehicle shows, ceteris paribus new innovations, more potential than using a dedicated freight vehicle. This is related to the lower rail operational costs. The transport of freight alongside passengers also reveals some potential, but this is only an option for small quantities of goods.

This doctoral dissertation adds to the existing body of knowledge in several ways. Firstly, the research makes an original contribution to scholarly theory. A unique characteristic of this research is the fact that the framework is developed for an urban rail freight context, and the viewpoint of the government is adopted. The key strength is the development of a generic tool which can easily be applied to different worldwide cases. Secondly, the developed tool provides policy makers the boundaries for an adapted policy towards for example subsidies. The results of this research show the benefits to society of a modal shift from road to rail. Thirdly, the framework helps private actors to understand under which conditions shifting from road to rail can become interesting for them.

On Applying Normalized Systems Theory to the Evolvability of Open Source ERP Systems - Ornchanok Chongsombut (16/09/2019)

​Ornchanok Chongsombut

  • 16 September 2019
  • Supervisor: Prof. Jan Verelst

Abstract

Enterprise Resource Planning (ERP) systems have become the most important part of enterprises’ information systems. Moreover, evolvability is widely considered to be an important concern for the design and development of software architectures. In particular, evolvability can be considered as a criterion to evaluate and analyse the quality and usefulness of ERP packages.

The purpose of this research is to investigate the evolvability of ERP systems and explore the potential for improvement in evolvability based on Normalized Systems theory.  More specifically, the research question is formulated as “How can the evolvability of open-source ERP systems be investigated and improved based on Normalized Systems theory?”.

To investigate this research question, the design science research approach is chosen as the methodology for the research. As a first step, four open source ERP packages were selected. Next, reverse engineering techniques were applied to the Sales module of each of these four ERP packages, in order to deconstruct their data and workflow models.  After exploring the data models of the selected ERP packages, the next stage involves the creation of prototypes based on Normalized Systems theory, which are then analysed in terms of evolvability based on a case study of fish market management in Thailand.  Finally, versions of the prototypes were generated for different technology stacks in order to illustrate their technology independence, which is an important aspect of evolvability.  This technology independence was not present in the existing fish market system, nor in the existing four open source ERP packages, and provides indications of the feasibility and advantages in terms of evolvability of the proposed approach.

This study has been one of the first attempts to redesign and redevelop parts of existing open-source ERP packages in the market by using Normalized Systems theory. Therefore, the dissertation can assist researchers in understanding the possibilities offered by Normalized Systems theory within their own domain of interest and help them explore the potential usefulness of the theory within each phase of creating an IT artefact such as analysis, construction, demonstration, and evaluation.

Practical contributions include that this dissertation resulted in four re-engineered data models of the Sales module of open-source ERP packages, which are available for practitioners for inspection, which could for example be useful in ERP package selection.  Additionally, this research provides insight into the feasibility into developing evolvable ERP systems by deconstructing existing open source ERP systems and regenerating them as Normalized Systems applications.

Investigating Non-Price Factors in Electricity Demand Response: An Applied Economics Study - Aman Srivastava (09/09/2019)

Aman Srivastava

  • 9 September 2019
  • Supervisors: Prof. Dr. Steven Van Passel and Dr. Erik Laes

Abstract

Even while the electricity sector is responsible for a large share (25%) of greenhouse gas emissions, the global demand for electricity is expected to keep increasing. While greater generation from renewable sources can ease both these concerns, such generation is variable in pattern, and the peak generation that is used to fill the gaps between electricity demand and supply is particularly costly.

One way to deal with the variabilities and costs of electricity supply is to match demand to it, through electricity demand response programs. These programs use mechanisms such as the time-based pricing of electricity, or restricting appliances usage at peak times, to shift electrical consumption to times when supply is higher and/or inexpensive.

While these programs have been implemented across many countries, particularly in the industrial and commercial sectors, they are less prevalent in the residential sector, and their success in this sector has been uneven, due to the limited flexibility of households. Further, such programs are not commonly used in developing countries, despite their potential for yielding multiple benefits.

This thesis uses a mix of three econometric approaches to explore how the induced flexibility in the electricity demands (a.k.a. responsiveness) of households may be improved. It draws upon historical cases from a number of different countries, and conducts forward-looking survey-based experiments in two of them – Belgium and India – to see how non-price factors such as socio-economic situations, demographic details, and individual attitudes affect the potential for flexibility and how these factors are valued.

Results demonstrate that demand response programs could indeed be successful, if they are designed and implemented carefully. For instance, they are more likely to succeed in urbanized regions with higher rates of economic growth. Some results were found to be common across the different country contexts. Factors such as household income, respondent age and gender, and awareness of the benefits of such programs, for example, influenced responsiveness to such programs across countries. Other results were specific to the local contexts. In particular, customers’ concerns about the privacy of their electricity usage data and their environmental attitudes influenced response in Belgium, while financial loss aversion and concerns about the stability of electricity supply were bigger concerns in India.

The shared and specific results offer lessons for both, international and domestic electricity policymaking and communications strategies, and point to areas for future research.

Essays on inequality, knowledge economy and financial development in Africa - Vanessa Simen Tchamyou (06/09/2019)

​Vanessa Simen Tchamyou

  • 6 September 2019
  • Supervisors: Prof. Dr. Guido Erreygers and Prof. Dr. Danny Cassimon

Abstract

The objective of this Ph.D. dissertation is to investigate how the four dimensions of knowledge economy which are consistent with the World Bank’s Knowledge Economy Index interact with financial access in order to mitigate income inequality in Africa.

The thesis consists of four empirical papers, notably: (i) the effect of information and communication technology (ICT) in complementing financial access to reduce inequality; (ii) the role of information sharing offices in modulating financial access to mitigate inequality; (iii) the impact of financial access in complementing education and lifelong learning to reduce inequality and (iv) the role of governance quality in modulating the effect of financial access on inequality in the African context. The underlying papers are capturing all the four dimensions of the World Bank’s Knowledge Economy Index (KEI), namely: education; ICT; economic incentives by means of information sharing offices; and governance quality or institutional regime.

The modelling exercise is based on the Generalized Method of Moments (GMM). Data are combined into a panel structure and sourced from: (i) the Financial Development and Structure Database (FDSD) of the World Bank for financial access variables; (ii) World Governance Indicators (WGI) of the World Bank for governance variables; (iii) World Development Indicators (WDI) of the World Bank for other variables and (iv) the Global Consumption and Income Project (GCIP) for income inequality variables.

The common denominator of the results is that, among the financial access indicators, the financial system deposit (or financial system depth) channel is fundamental in mitigating income inequality by means of the engaged knowledge economy policy instruments.

Policy and scholarly recommendations are proposed.

Unlocking the benefits of pick-up points for sustainable E-commerce distribution in urban areas - Ivan Cardenas (23/08/2019)

​Ivan Cardenas

  • 23 August 2019
  • Supervisors: Prof. Dr. Thierry Vanelslander and Prof. Dr. Wouter Dewulf

Abstract

E-commerce has been increasing dramatically during the last decade. This disruption in shopping behaviour has brought about changes to the way freight moves in cities. For example, in the US, in 2009, one out of three freight trips had a household as a final destination. Within only 10 years, by 2019, the number of household deliveries had tripled and now three out of five freight trips are destined to households. For the future, this trend is expected to continue.

Despite the obvious convenience of home deliveries, the last mile of e-commerce is the most expensive leg of the supply chain. Additionally, home deliveries cause a number of negative impacts to residential areas such as street blocking because of double parking, an increased risk of accidents, as well as other negative externalities linked to freight vehicles (e.g. pollution, congestion, noise).

Pick-up points are therefore widely regarded as a solution for increasing the efficiency and reducing the traffic of freight vehicles in residential areas. If doorstep deliveries get consolidated in pick-up points, household freights trips become less frequent. Municipalities and companies are recognising this potential and start focusing their attention on densifying pick-up point networks. From their perspective, those pick-up point networks constitute an instrument to mitigate the negative impacts of e-commerce in urban areas. Estimations suggest that in Europe so far there are about 200,000 pick-up points, while in Belgium this number is getting close to 7,000.

Simply densifying the pick-up points may however lead to unintended consequences: the positive side is that a high density of pick-up points provides more convenience to the customer while at the same time discouraging the use of individual motorized trips to collect packages. However, if there are too many pick-up points, the consolidation effect gets lost commercial vehicles will still need to drive in residential areas causing high operational and external costs.

In this dissertation, the conditions under which pick-up points can be beneficial for society were disentangled. A cost model was developed to understand the effects on operational and external costs resulting from location and density patterns of pick-up points. Our results show that the total costs (external and operational) are almost insensitive to the density of the pick-up point network. However, it appears that the same costs decrease proportionately to an increase in the flow of parcels through the pick-up point network. This suggests that a wider acceptance for off-home deliveries will further unlock the benefits of pick-up points. To reach a larger audience, not only geographical proximity between home and pick-up points should be considered, but also other service level attributes.

Retail logistics costs and policy impact - What is the total cost to secure innovation for a greener retail supply chain? - Konstantinos Papoutsis (12/07/2019)

​Konstantinos Papoutsis

  • 12 July 2019
  • Supervisors: Prof. Dr. Eftihia Nathanail, Prof. Dr. Thierry Vanelslander and Prof. Dr. Wouter Dewulf
  • Joint defence UAntwerp - School of Engineering (University of Thessaly, Greece)

Abstract

Public-policies on freight mobility record tangible cost impacts on supply chains and in logistics processes such as goods handling, transportation, etc. Especially for the retail sector which requires a more responsive performance, the effects are even more critical. Thus, society is impacted by air pollution, noise nuisance, etc. for which they are not accountable, and decision-makers are trying to tackle via new policies. Therefore, sustainability has a cost that should be paid by polluters. On the contrary, retailers seek for innovation to offset the additional operational implications. A holistic approach on pinpointing and estimating the total costs that retailers pay in order to ship their products to customers embedding the policy adoption cost is the main scope of the doctoral dissertation.

A cost model of a retail supply chain is built focusing on urban context and the effects of policies and innovations on total costs and sustainability performance of a retailer’s B2B supply chain are estimated. This is achieved by conducting an Activity-Based-Costing method, identifying logistics activities and resources needed to execute them and incorporating public policy cost elements into this analysis. With view to internalize the external costs generated, the transport externalities are also integrated together with the indirect cost of delay in delivering time-sensitive products such as FMCG, etc.

To understand the actual costs of deliveries the cost model is applied onto the operational model of a food retailer delivering to their outlets in the city of Antwerp. On top of the as-is scenario, three different delivery scenarios are examined (urban consolidation center - UCC), Tethering, shared bus) to investigate the total cost effect of mobility policies on retailers’ supply chains. UCC secures significant cost savings but this is subject to a flawless consignment and cooperation between stakeholders. External costs constitute integral part of sustainability and thus, a sustainability analysis is performed between scenarios, encompassing service and social perspectives. In that regard, Tethering is the most sustainable scenario due to the shorter distance covered by trucks in one roundtrip.

This dissertation explores the potential of retail logistics both from a policy-making and industry point of view. It portrays the dynamics and drivers of a key sector and it reveals similarities between different economic sectors: insights could be transferable accords sectors. The potential room for innovation adoption is indicated by quantifying the cost of innovative scenarios to the retailers. It, finally, crystallizes how ‘the polluter pays’ principle is blended into the retail logistics costs.

Antecedents and Consequences of Complex Costing Systems: Empirical Analyses - Piyada Daowadueng (08/07/2019)

​Piyada Daowadueng

  • 8 July 2019
  • Supervisor: Prof. Dr. Ann Jorissen

Abstract

During the last few decades many organizations in both the manufacturing and service sectors have been faced with dramatic changes in their business environment, for example deregulation, increasing global competition, the reduction in product life-cycles arising from technological innovations, and customers’ demands for greater product diversity. These changes have had a significant influence on costing systems design; that is, these changes have led to modifications of cost structures and cost behavior in which traditional costing systems fail to provide the relevant information for companies. As a result, there is an interest and a need to adopt more complex costing systems that can provide more detailed cost information that is both useful and relevant for decision-making, performance measurement and control.

Meanwhile, focusing only on a level of cost allocation detail cannot distinguish sufficiently between the different levels of complexity in costing systems that exist across all companies. In our study, we examine complex costing systems by considering both the level of cost allocation detail and the variety of different costing systems being used (standard costing, variance analysis, contribution margin, life-cycle costing and target costing) in companies.

The objective of this dissertation is twofold. First, we want to investigate the antecedents of the adoption of complex costing systems (cost allocation detail and costing systems diversity), and second, we want to examine the consequences of complex costing systems. To address these two research objectives, we provide three empirical studies.

The first empirical study, we investigate that the presence of internal uncertainty (i.e., operational turbulence) is correlated with the adoption of complex costing systems. Next, the second empirical study, we argue that managers will use more detailed and more diverse cost information to support decision-making and decision-influencing control practices through their perceived information accuracy. Lastly, the third empirical study, we examine the extent to which cost information is used in exploitation or exploration firm context in a decision-making and decision-influencing role, hereby we investigate whether the level of detail and diversity of cost information present, moderates this relationship between an exploitation or exploration context and the use of cost information in a decision-making or decision-influencing role. To test hypotheses, we collected data from 191 medium- and large-sized Thai manufacturing and service companies.

Speculative Asset Prices - Gertjan Verdickt (26/06/2019)

​Gertjan Verdickt

  • 26 June 2019
  • Supervisors: Prof. Dr. Jan Annaert and Prof. Dr. Marc Deloof

Abstract

In this thesis, I tackle several key questions in financial economics. First, I focus on the question “why do stock prices move over time”? In theory, stock prices should equal the expected discounted cash-flows. However, does the variation in stock prices come from changes in expected cash-flows or discount rates? Why is this important? From the standpoint of finance practitioners, understanding the risk-return relationship can improve tactical asset allocation decisions and is crucial to assess the potential of new projects. For academics, evidence in return and dividend growth predictability disputes classic asset-pricing theories.

Second, I address the question if managers adjust important decisions, such as dividend policy or the timing of initial public offerings, when the threat of war increases. For long-run investors, dividends form the largest part of the income. Having an understanding of managerial behavior, in particular when the disaster probability increases, is important for asset allocation decisions. Moreover, I test whether changes in stock prices in time of increased military conflict are driven by changes in expected cash-flows or changes in discount rates. In general, I show that time-varying expected returns and expected dividend growth are essential to understand stock price changes.

Struggling between control, competition and collaboration: concepts about hybridity and its main antecedents and consequences - Lode De Waele (17/06/2019)

​Lode De Waele

  • 17 June 2019
  • Supervisor: Prof. Dr. Arjen van Witteloostuijn​

Abstract

Public organizations attempt to cope with processes of institutional change by continuously looking for ways to ensure legitimacy as their survival - in contrast to their profit-driven counterparts within the private sector - is largely dependent on the public support obtained from elected officials, civil servants and civilians. Yet, the demands stemming from the institutional environment are diverse and often reflect conflicting viewpoints. Public organizations in response to these challenges employ mechanisms such as selective coupling in an attempt to balance between conflicting institutional logics, which cause the emergence of hybrid forms.

However, literature lacks clear conceptual frameworks that define how these institutional logics actually look like. Furthermore, within the current discourse in the field of Public Administration, empirical research addressing the main drivers of hybridity is inadequate, resulting in a limited understanding of its primary antecedents. Finally, although hybridity has been found to potentially affect an organization’s performance, there is scant empirical research examining these claims.

Relying on a neo-institutional framework, this dissertation identifies and defines the main institutional logics within public organizations, studies which processes drive public organizations towards hybridity, and analyzes how hybridity might affect organizational performance. Some overall conclusions are that core institutional changes drive public organizations toward developing hybrid forms; yet, public organizations in response to these changes tend to pursue different strategies, ranging from segmentation to segregation and absorption. Furthermore, hybrid forms are likely to cause considerable performance-tensions.

One of the most challenging performance-tensions has been empirically confirmed, using a quasi-experimental research design. As a result, we found that public servants with high levels of Public Service Motivation in hybrid organizations that combine bureaucratic and New Public Management-related logics are more likely to engage in acts of pro-social rule-breaking in order to pursue customer-oriented goals, which results in client-based discrimination. In conclusion, although not every organization is subject to hybridity, our studies showed that hybridity is a widespread phenomenon, caused by large shifts in institutional demands with substantial effects on organizational performance.

Optimisation of chemical batch plant design: mathematical models and algorithms for strategic decision support - Floor Verbiest (14/06/2019)

​Floor Verbiest

  • 14 Juni 2019
  • Supervisors: Prof. Dr. Trijntje Cornelissens and Prof. Dr. Johan Springael

Abstract

Food additives, pharmaceuticals and lubricants are, among other products, typically manufactured in chemical batch plants. The construction of such batch plants comes with significant investments, which require appropriate capacity assessments. Such assessments, or strategic capacity decisions, are the main topic of the strategic batch plant design problem. This problem entails determining the optimal number and size of equipment units for every production stage, as well as the optimal operational planning guidelines, so as to minimise total costs while satisfying both demand and design related constraints. These demand and design constraints generally state that the designed plant should be large enough to produce a given, total demand for a range of products within a given production horizon.

The batch plant design problem has been studied extensively in literature over the past four decades and is usually formulated as a mixed-integer linear programming problem that is solved exactly. There is, however, still a need for bringing the existing models and solution techniques a step closer to reality. With this goal in mind, a first contribution of this dissertation relates to the modelling of real-life chemical batch plants. Such plants typically consist of parallel production lines, i.e. lines that can be dedicated to particular products or product families, which we included as a design option into existing plant design models found in literature. Secondly, we investigated the integration of the operational use, i.e. how the plant is used, into the plant design problem with parallel production lines. Apart from a single product campaign mode of operation used earlier, two additional, more complex, modes were examined: mixed-product campaigns and network planning.

As a third contribution, different solution techniques, such as a matheuristic optimisation algorithm, were developed to solve these design problems with parallel production lines within limited computation time. Finally, we considered the multiperiod batch plant design problem and proposed a multiperiod delivery framework. This framework allows to systematically combine different delivery and production planning characteristics that arise in such a multiperiod context. We performed an exploratory study to investigate the impact of this multiperiod delivery framework on the plant design, and this for both a plant operating in a single and a mixed-product campaign mode.

Essays on Servitization - Wim Coreynen (13/06/2019)

​Wim Coreynen

  • 13 Juni 2019
  • Supervisors: Prof. Dr. Paul Matthyssens, Prof. Dr. Arjen van Witteloostuijn and Prof. Dr. Johanna Vanderstraeten

Abstract

Servitization describes the addition of services to manufacturers’ core product offerings to create additional customer value. This thesis contributes to the servitization research field through four empirical studies and one tool for practice.

The first essay 'Boosting servitization through digitization' is concerned with the question how technology enables firms to offer better value to customers through services. In this essay, we explore different pathways for digitally-enabled servitization and apply the dynamic resource-based view to pinpoint specific resource configurations at different levels of dynamicity supporting each pathway.

In the second essay 'Internal levers for servitization', we look into service upscaling from three perspectives, namely a content-, process- and context-perspective, to explore the organizational barriers that inhibit firms from servitization. We also develop and test a new methodology (consisting of a diagnostic instrument and set of workshops) to help firms upscale the service business.

The third essay “The antecedents of digital servitization” revisits the topic of technology-enabled servitization addressed earlier. Here, we use a dynamic capabilities perspective to investigate the relationships between firms’ learning mechanisms and the likelihood of them developing a strategy for servitization, digitization, or a combination of the two, in a deductive fashion. In addition, from a contingency perspective, we consider the influence of the environment as a potential amplifier of these relationships.

In the fourth essay “What drives servitization?”, we move into rather uncharted research territory: i.e., the microfoundations of servitization. One of the elements contributing to servitization success (or failure) is the motivation of the decision-maker(s) in charge of the service transition process. Contrary to the third essay, we use an abductive approach (the purpose being to develop theory rather than test it) and apply a behavioral strategy lens to explore the motivational and strategic drivers of decision-makers for servitization.

The final essay “Are you ready for servitization?” is not an empirical study as such and therefore added as an appendix chapter at the end of the thesis. In this essay, we offer an assessment tool to be used by practitioners to measure firms’ capacity for servitization. First, we zoom in on several aspects identified throughout the thesis that play a crucial role in the successful transition towards services. Next, we present the tool and discuss the results of two case firms that have applied it.

African Eurobonds: opportunities and challenges for sustainable economic growth after the HIPC initiative - Christian Senga (29/05/2019)

Christian Senga

  • 29 May 2019
  • Supervisor: Prof. Dr. Danny Cassimon

Abstract

Since 2006, countries from Sub-Saharan Africa (SSA) have been one after the other taping international capital markets to mobilize substantial funds through sovereign and government-guaranteed eurobonds. To the surprise of many, these eurobond issues have almost always been oversubscribed, thus indicating a high appetite of international investors for these securities. This investors’ enthusiasm is indeed puzzling since, just some years ago, this region was mired into a severe sovereign external debt crisis that could have not otherwise been solved but through debt forgiveness in the framework of the Heavily Indebted Poor Countries (HIPC) initiative by the International Monetary Fund and the World Bank.

Eurobond issuers and their proponents justify this move by the need for these countries to diversify their sources of development funding and, thereby, reduce their dependence on the already drying up foreign aid from developed countries. It is also seen as an opportunity for them to register on the investors’ radar and attract more foreign direct investment. However, some critics find this investors’ enthusiasm rather justified by their thirst for high yields outside their post-financial crisis domestic environment dominated by protracted low interest rates and sluggish economic recovery. They draw parallels between the current situation and the reckless lending of the 1970s that led to the above-mentioned debt crisis in developing countries, and warn about the possibility of its resurgence in SSA should this eurobond spree spiral out of control.

This thesis endeavors to tackle this important concern by investigating the sustainability of the SSA eurobond market, as well as its potential for economic growth in the issuing countries. It first scrutinizes the drivers of these bonds’ secondary market yields to grasp the incentives for the quality of macroeconomic management in the borrowing countries. Then, it explores the avenue of international portfolio theory to evaluates the possibility of diversification benefits that may provide alternative justification to the interest of international investors in SSA eurobonds. Finally, it investigates whether and how government borrowing through international capital markets affect investment dynamics in these countries.

Dividend policy: a long-term investigation - Leentje Moortgat (20/05/2019)

​Leentje Moortgat

  • 20 May 2019
  • Supervisors: Prof. Dr. Jan Annaert and Prof. Dr. Marc Deloof

Abstract

The return from investing in stocks can be disentangled in two components: capital gains and dividends. For long-run investors, dividends form the largest part of the income. In this dissertation, we investigate dividend policy of firms listed on the Brussels Stock Exchange using data of almost two centuries (1824−2012).

In the first study, we investigate how dividend policy evolves over our sample period. We show the institutional environment in which firms operate, which is generally believed to affect corporate dividend policy, changed drastically: investor protection improved seriously over time and dividend taxation was introduced. Despite these huge changes, we find dividend policy to be surprisingly stable over time.

The second study explores whether the motives to pay dividends depend on (1) the institutional environment in which firms operate, (2) the state of the economy and (3) the maturity of the industry in which the firm is active. We find common determinants of dividend policy in different environments. Large firms, firms with a low level of idiosyncratic risk, firms with a high share price and more liquid firms are more likely to pay dividends. While this is consistent with life-cycle arguments, our evidence raises doubt about signaling as a first-order explanation of dividends.

The third empirical study adds a new explanatory variable to the dividend debate. Consistent with imprinting theory, we show that dividend policy at the moment of the firm’s first listing has a long-lasting impact on future dividend policy. However, the impact of the initial dividend policy decreases when a company goes through its lifecycle. This indicates managers are both firm and flexible when setting their dividend policy.

In summary, stability seems to be the common denominator. Study I shows dividend policy is very stable over time. Next, we find that the determinants of the decision to pay are stable in different environments. Finally, we show that dividend policy has a stable component that persist over a company’s life.

Health economics of vaccine preventable diseases in selected low and middle-income countries - Carlos Lara Oliveros (21/03/2019)

​Carlos Lara Oliveros

  • 21 March 2019
  • Supervisor: Prof. Diana De Graeve

Abstract

From a health economics perspective, this thesis presents evidence-based considerations for decision-making, including the quantification of the economic and disease burden, the economic evaluation of vaccine intervention strategies, and the analysis of determinants of the supply and demand for vaccines. More specifically, it contributes to the body of knowledge of the health economics of vaccine preventable diseases in low and middle-income countries (LMICs), for which there are important knowledge gaps. The aims of the thesis are to increase the body of knowledge on LMICs with special attention to Colombia, and to propose explanations for the continued high economic and disease burden despite efforts to facilitate access to vaccines especially for vulnerable populations.

We also contribute to the existing literature by improving the estimations of the related costs per episode of acute respiratory infections and by generating input parameters for future studies. For the specific case of Colombia, we recommend to the national health authorities improvements in the timing of the administration of the influenza vaccine. In addition, we provide national authorities with some points for consideration when deciding the introduction of a new vaccine by performing cost effectiveness analyses for Albania, Colombia and Egypt. We also elucidate the vaccination determinants for LMICs with a subsidized health care system especially in the case of Colombia. We therefore recommend to health authorities to consider the inclusion of targeted programs in their vaccination strategies. Overall, we hope this thesis demonstrates that decision-making for the optimal allocation of resources for strengthening vaccination strategies is complex and should consider many perspectives.

The logistics sector in a consumer driven society, essays on location and network structure - Joris Beckers (18/03/2019)

Joris Beckers

  • 18 March 2019
  • Supervisor: Prof. Dr. Ann Verhetsel

Abstract

Over the past twenty years globalization forces have internationalized production and consumption. The global disintegration of the production process severely complicated the in- and outflow of materials and significantly expanded the operational scale of logistics companies. Combined with an increasing awareness for the consumer, with a faster delivery of qualitative goods, this global scope demanded close cooperation between producers, suppliers and distributors and gave rise to a strategy of supply chain management and the advent of contemporary logistics. Recently, the popularity of the internet as an extra retail channel further complicated the organization of logistics flows. The emergence of free home deliveries implied a fragmentation of both destinations and volumes and constituted the consumer as an inherent part of the distribution process.

The changing role of logistics implied new locational factors for the physical distribution processes and caused a spatial reconfiguration of the logistics network. By using four case studies, this dissertation studies how local spatial interactions shape this new spatial configuration of the logistics network in Belgium. The first two cases engage in network methodologies to quantify this objective for the whole logistics sector. They confront network-based insights gained from a microeconomic dataset of buyer-supplier linkages with location- and domain-specific insights. The last two case studies focus on a subset of the sector: the courier, express and parcel (CEP) services. They highlight the impact of the consumer on the logistics chain by means of a mixed method methodology.

The network analysis identifies the presence of spatially contiguous communities within the Belgian logistics networks. Contrasting these with indicators of employment concentration results in a typology of logistics clusters. These communities are later refined by iterating the community detection algorithm, which uncovers the hierarchical structure within the logistics buyer-supplier network. Next the third case study maps the geography of online shopping behavior in Belgium. This lays the foundation for an assessment of the spatial distribution of the demand for online ordered parcels and leads to the development of a framework of new infrastructures in e-commerce delivery chains.

Drawing inspiration from network methodologies proves to allow for a richer description of the geography, especially with the increasing availability of ever growing datasets. Its application demonstrates how interactions differ geographically and hence impact the spatial configuration of the logistics network on a local level. However, this logistics network is not static. The rise in online shopping, with the demand for ever faster and more customized deliveries by consumers, extended current CEP networks with a new urban layer of logistics infrastructures. While this is currently a very fragmented layer in comparison to the other steps in the distribution chain, a clustering hereof is necessary to accomplish a sustainable organization of the delivery of online orders.

Food inaccessibility in Flanders: identifying spatial mismatches between retail and residential patterns - Jeroen Cant (06/03/2019)

Jeroen Cant

  • 6 March 2019
  • Supervisor: Prof. Ann Verhetsel

Abstract

This dissertation explores whether food retail inaccessibility has established itself in Flanders. To investigate this a mixed methods methodology is applied. First, a theorization on how inaccessibility can occur even in highly developed nations is provided, based on a comprehensive literature review. Historically retail has been embedded in residential structures. Monofunctional development patterns in Fordist and post-Fordist times, however, have led to a severe unbundling of living and shopping, resulting in an increased need for automobility. Those with reduced personal mobility rates due to socioeconomic or physical reasons are confronted with inaccessibility because of this spatial mismatch.

Retail policy, however, has been shown to temper unbundling trends, particularly when it is embedded in spatial planning and applied on a national/federal level. In Flanders, policy was based on socioeconomic restrictions with the municipalities in the driver’s seat. Descriptive statistics then show important retail-residential unbundling trends in the region. The relationship between food retail floor space and the residential is further investigated using Geographically Weighted Regression (GWR).

Large supermarkets and hypermarkets have clearly sprawled away from residential structures, though the network of traditional, specialized retailers is still embedded. Big shops, however, increase consumer utility through lower prices and larger assortments. An address level network analysis shows that basic accessibility, or the ease of reaching a food opportunity with a full variety regardless of attractiveness, has been maintained in cities and in town centers through a network of both traditional stores and small supermarket formulas.

Since poverty has sedimented in the urban core, basic inaccessibility due to socioeconomic reasons (i.e. food deserts) remained limited and highly contextualized in outer city modernist housing estates, garden cities and industrial areas. Said small food store network never established itself in the affluent monofunctional neighborhoods that characterize urban sprawl in Flanders, leading to high local travel distances. As historically personal mobility rates were high here, actual inaccessibility remained limited for a long period of time. Currently, however, these areas are confronted with important aging trends, and issues will start manifesting themselves. Given the prevailing socio-spatial evolutions, both the unbundling of large-scale food retailing and the residential, and growing inaccessibility in the suburbs are expected to rise further in the near future if not in some way combated.

Evolution of entrepreneurial teams in technology-based new ventures - Tatiana Zabara (18/02/2019)

Tatiana Zabara

  • 18 February 2019
  • Supervisors: Prof. Dr. Arjen van Witteloostuijn and Prof. Dr. Christophe Boone
  • Joint defence UAntwerp - Tilburg University

Abstract

Technology-based entrepreneurship is an influential source of scalable economic growth and major improvements in public health, environmental sustainability, and wealth creation. The mechanisms through which entrepreneurs shape their ventures are often contingent on factors related to the institutional characteristics of the national economy, industry, and most importantly – the entrepreneurs themselves.

This PhD dissertation, highlights the essential role of the entrepreneurs at the heart of technology-based new ventures – with a specific focus on entrepreneurs’ experience, disposition and actions. This dissertation consists of three papers: one literature review and two empirical studies. First, a systematic literature review was performed to map extant work and to identify existing gaps and avenues for future research. We then aim to address these research gaps in a unique longitudinal dataset of Flemish technology-based new ventures and founders’ career histories.

The first empirical study examines the role of lead founder’s personality traits as a source of potential bias in assembling and structuring the founding team. By delving into the micro-foundations of founding team structures and focusing on the entrepreneurs’ individual disposition, this study aims to contribute to the important question of why management teams look the way they do and why there is high degree of heterogeneity with regard to how founders start their firms.

The second empirical study examines the antecedents of a new managerial hire in technology-based new venture. We argue that antecedents of new member addition can be traced to the attributes of team, organization and environment in which new ventures operate. We further examine and discuss the relative importance of these attributes. Overall, this dissertation provides a deeper understanding of the role of individuals and teams in new venture creation, with a specific focus on team emergence and evolution.

Data mining for tax fraud detection - Jellis Vanhoeyveld (14/02/2019)

Jellis Vanhoeyveld

  • 14 February 2019
  • Supervisors: Prof. Dr. Ir. David Martens and Prof. Dr. Bruno Peeters
  • Department of Engineering Management

Abstract

Due to the impact, reach and diversity of tax fraud, governments invest in advanced detection methodologies. Data mining techniques offer an important opportunity as they can automatically distinguish fraudulent patterns from legal ones and tax administrations can subsequently focus their limited resources on the most likely presumed fraud cases.

First, we focus on the imbalanced data distribution problem for behavioural data, where a variety of tailored imbalanced learning solutions are developed and examined across imbalanced behavioural datasets arising from different application areas. Note that the tax fraud detection domain is also characterized by imbalance, where the number of compliant cases severely outnumbers the number of fraudsters and this causes suboptimal performances for many supervised classification techniques. Furthermore, behavioural data that capture the fine-grained (inter)actions of persons and/or organizations are also widely available (e.g. invoicing data in the VAT domain, people appearing in the board of directors of organizations, shareholder ownership, etc.), though they are mostly ignored in tax fraud detection studies. Next, we leverage the previously obtained insights in a case study of customs fraud. Our main findings suggest that fine-grained behavioural and high-cardinality data (e.g. consignee, declarant, type of commodity, etc.) are very predictive and can even outperform traditional data that are used in the literature and contain many more variables. Furthermore, imbalanced learning solutions can further improve the predictive performance of the models.

Unsupervised anomaly detection constitutes another class of data mining methodologies. It involves the detection of entities (fraud cases) that show a conduct that differs significantly from normal behaviour (compliant cases) solely based on their feature representation. As the number of tax declarations can become large (e.g. several millions), many anomaly detection techniques are too computationally involved to be applied in such settings. We develop efficient data compression techniques as a pre-processing step so that the detectors are presented with much fewer (though relevant) cases. In a case study of VAT fraud, such scalable anomaly detectors are applied to the feature representations of all companies pertaining to the same sector. Based on domain knowledge, we derive presumed fraud indicators (features) that take the form of tax ratios from the VAT declaration form and client listings.​

Operations research to the rescue: Increasing disaster preparedness by pre-positioning emergency supplies - Renata Turkeš (28/01/2019)

​Renata Turkeš

  • 28 January 2019
  • Supervisor: Prof. Dr. Kenneth Sörensen
  • Department of Engineering Management

Abstract

Hundreds of millions of people suffer yearly as a result of natural and man-made disasters. No country is immune from the risk of disasters, but much human loss and suffering can be avoided by preparing to better deal with these emergencies. A common mechanism of increasing disaster preparedness is to pre-position the emergency supplies (such as water, food or medicine) at strategic locations, so that the aid can be distributed to affected areas immediately in the early post-disaster.

Humanitarian logistics is a critical element of an effective and efficient disaster relief process: planning a pre-positioning strategy requires deciding on the number, location and size of storage facilities to be open, the quantities of various types of emergency supplies to be pre-positioned at each facility, and the distribution of the supplies to demand locations after a disaster, under uncertainty about demands, survival of pre-positioned aid and transportation network availability. Operations research has the potential to help relief agencies rescue lives and maximize the use of limited resources by optimizing emergency strategies, but also the potential to help the field of humanitarian logistics to mature further.

One of the most vexing issues in the research on pre-positioning and other humanitarian logistics problems is the lack of data. With the aim of making the first step in assembling a set of benchmark instances for the pre-positioning (and related) problem(s), we generated and shared 30 diverse case studies, and a random instance generator that can be used to construct arbitrarily many various instances of any size. Another important issue that we have identified is the common mathematical model that includes intangible and controversial costs for unmet demand. We provide both theoretical and numerical evidence that shows that there exists an alternative mathematical formulation which obtains the same quality of emergency strategy without any performance loss.

Since the pre-positioning problem is NP-hard, we developed a matheuristic optimization algorithm which is able to solve large instances within limited computation time. An extensive computational study that employs the developed framework (problem library, mathematical formulation and solution algorithm) helps us to gain insights about the influence of a number of disaster properties and their interactions on the pre-positioning planning decisions. We use this knowledge about the problem to also bridge theory and practice by deriving trustworthy guidelines which humanitarian workers can use on the ground to mitigate the devastating impacts of new potential disasters.