Examples of Master projects 2017

Summary:
With the growing popularity of sustainability in the business sphere and the concept of corporate social responsibility, there is increasing need for companies to aim for the contribution to sustainable development in a formal way, making use of tools that, when used properly, can contribute to the effective overall governance of sustainability issues. These tools are mainly—but not exclusively—technological. In this project, we aim to understand the current state of the perceptions around corporate social responsibility, to properly situate Yanstra—an in-house developed tool—as a social service, verifying the suitability of this tool in the corporate social responsibility domain.
The suitability of Yanstra as a social service was analyzed in the context of a general case study: water management issues in the sugar segment in Uttar Pradesh, India. Some of the most representative relationships between relevant stakeholders in this segment were modeled using some methods from the Systemic Enterprise Architecture Methodology (SEAM), a methodology for strategy thinking developed by the Laboratory for Systemic Modeling at EPFL. The service was placed in the models to determine what types of value it can potentially bring to the underlined stakeholders, corporate and societal.

Results:
The main contributions of this project are the models from the case study and a classification of the different situations encountered in the corporate social responsibility domain, along recommendations on what methods from SEAM are most adapted to analyze them.
Two main relationships between stakeholders in the case study were modeled: farmers with the local governmental irrigation department and beverage companies with farmers and local communities. While it was seen that the two relationships have potential to be enhanced with the adoption Yanstra, for the moment it would be better to pursue its introduction of in the farmer-irrigation department relationship, as further features would need to be developed for the service in the context of the other relationship to achieve the desired outcomes that Yanstra has as a service offering.

Summary:
Based on the global strategy plan of ABB, three supply chain management organizations have been merged in Switzerland, Zurich Oerlikon. This consolidation demanded a harmonization of business processes in the newly formed organization. The objective of this thesis is to analyze existing business processes and to identify consolidation opportunities for predefined teams. Thus, new consolidated to-be business processes have been defined and visualized for each team. To achieve this, appropriate business process management procedures have been applied. Concurrently, crucial change management aspects have been considered to attain a successful result. A structured procurement landscape was designed additionally to achieve an overall overview. The increased transparency, with clearly defined team-interfaces, enables to foster the future cooperation between the teams in the new supply chain management organization.

Summary:
Today, agile software development is growing in popularity. Smart and connected products offer expanding opportunities for new capabilities. They raise a new set of strategic choices about how industries should work to develop these capabilities and capture value. For industries like pharma, aviation, food, and tobacco, where software design and development was not a core element, connected products bring many challenges such that building the right IT infrastructure is critical to enabling new capabilities and unlocking their business value. In order to develop these complex products, many organization implement customized agile methods tailored to better accommodate their needs. Initially, agile methods were seen as best suitable for small and collocated teams. However, very few studies show the applicability and the suitability of agile methods in geographically distributed development.
This qualitative case study was conducted to investigate how scrum was newly adopted in the context of distributed software development. This study was exploratory and focused on one product development of a complex program at Philip Morris International (PMI) in the Information Services department. The data was collected from interviews, meetings, workshops, training, face-to-face conversations and phone calls. Focusing on seven challenges that were identified through a literature review on the topic, the principal aim was to assess, illustrate, and discuss how agile methods were implemented in this context and how agile transformation is taking place.

Results:
This study is a response to the frequent calls for more industrial case studies on agile implementation in global software development. This case study addressed this research gap. It examined the way Agile was implemented in PMI focusing on one development team using Scrum. The case study included twenty-seven interviews, workshop observations and phone calls. The study has taken the form of a comparative analysis. Existing challenges identified in various case studies and literature reviews included traceability, requirements engineering, quality insurance, hierarchical management & organizational boundaries, investment, coordination and communication and change resistance. By comparing those seven challenges, the findings showed that there are still various gaps between agile theory and the actual case showing an overall incomplete transformation. However, this study showed that transformation is coming from the bottom; people who are prepared to discard preconceptions, willing to try new untested approaches and bring initiatives. One of the lessons learned presented in this paper is the importance of mindset which showed helped to gain acceptance of the transformation. Including the right people will encourage other people to participate and improve the visibly of work in the new agile way and will make the transition. In addition, this study proposed a set of recommendations in order to fill the gaps; such as team collocation with on-site developers to solve the gaps.

Summary:
Today’s world has never been so uncertain and is evolving quickly. Industries must reinvent themselves all the time to keep up with consumers’ expectations and survive. Philip Morris International is undergoing the same kind of problem with the progressive decline of cigarette. It has therefore forced them to launch in 2014 a new product portfolio called Reduced-Risk Products. But bringing new products to market raises many challenges from a supply chain perspective. The existing processes must be rethought to take into account the new products intrinsic characteristics as well as the challenges brought by the 21st century. In this regards, information systems have always been considered to be classical key factors of success, especially in the supply chain field. At a time when data has become the new oil and company are being transformed by digital technologies, the possibilities are multiple.
The objective of this thesis is to understand how a big company such as PMI can use these new tools to go through its new business journey from a supply chain perspective.

Results:
An analysis of PMI supply chain current challenges was performed via interviews. The RRP portfolio success, embodied by the electronic heat-not-burn tobacco product IQOS, has surprised the company who cannot rely on its traditional supply chain anymore to support the growth of its business. Projects have already been launched and demonstrate PMI is going in the right direction. However, it will take time for the company to catch up and this thesis contains suggestions for improvement.

Summary:
In this paper, we studied the strategic positioning of the African only tech unicorn Jumia to face the inevitable arrival of the web giant on the African continent. We used a Resource Based View and the VRIO framework combined with case studies to assess the resources and capabilities which can provide a sustainable competitive advantage. The sustainability of the first mover advantage being challenged, we also provided strategic recommendations to further increase the long term profit of the company. Finally, we used a system dynamic modelization to verify quantitatively our qualitative analysis.

Results:
We have seen that the resource which can provide a sustainable competitive advantage is the knowledge of local specificities. Jumia leverages this knowledge by pursuing a differentiation strategy focusing on customer service quality. The sustainability of this advantage is challenged due to different reasons such as the turbulent environment and the market power of the giants which can sustain time compression diseconomies. Therefore, to stay ahead of potential competitors, Jumia should further provoke learning situation by launching experimentation in new countries and new industries. It will also enable the company to better leverage and monetizing its existing advantage, which is the knowledge of market and consumers.

Summary:
The objective of this work is to explore quality on a flow that is very particular to Rolex: the recycling of waste. Production workshops generate several types of waste and classify them into two categories: rich waste and undefined waste. For each type, a specific treatment is internally realized to recycle the metals. Rich waste goes through the casting grains sub-flow whereas undefined waste is burnt.
The goal is to implement a quality system on the overall recycling flow in order to measure its quality of service. Today, a quality system does exist but is implicit. For example, the team is able to detect if there is a defect during the process thanks to its experience. Visual parameters are used to catch irregularities.
Until now, this implicit quality system was enough. However, it is time to improve it. The goal is to reduce the gap between what the workshops are expected and the actual performance of the recycling flow. To do so, it is necessary to quantify the anomalies to find some room for improvement over the flow. It was decided to create a tool that is going to be able to measure this quality: an indicator.
First, this work contains an exhaustive section on available literature on quality. On the one hand, it aims to agree on a unique definition of quality. On the other hand, it presents quality systems as a whole and the main challenges behind the ISO 9000 certification.
Then, this work details the concrete implementation of quality within a company. It suggests an analytical approach to measure quality. Analyses and physical tests were put in place in order to develop a follow-up indicator. This tool serves two purposes: communicate with customers (i.e. production workshops) and identify areas of improvement internally. To conclude, collaboration with workshops will permit to take actions on suggested enhancement in favour of a better quality of service.

Summary:
As an analyst at C4 Ventures, I spent most of the time analysing new investment opportunities. In order to correctly assess each opportunity, it is important to be knowledgeable about the industry in which the start-up is evolving. This is why C4 Ventures has specialised in three themes: smart hardware, future of commerce and digital media. Regarding the smart hardware sector, a recent fast moving industry has emerged in the last decades: drones. The aim of my thesis was to understand this industry and be aware of the innovations and investment opportunities within this market.

Results:
To do so, I have analysed this market and its trends to create an up to date market map. I understood that the drone market has evolved from military purpose to a B2C market in early 2010 thanks to the emergence of smartphones which decreased the cost of several key components also present in drones. Shortly after, the B2B market emerged which represents a higher potential in terms of revenue. Indeed, drone technology is about to disrupt several industries such as the agriculture, construction, real estate, inspection, and much more. I havethen analysed two prior successful drone start-ups coming from EPFL: senseFly and Pix4D. With all this material, I have then analysed and shared start-ups I consider could be great investmentopportunities. These opportunities are innovative solutions to resolve some of today’s drone industry problems such as safety concerns, regulations concerns, improving automatization, and the verticalisation of this technology to a specific industry like the construction industry.

Summary:
In the context of its Digital Transformation program, Nestlé has many on-going digital projects. This master thesis focuses on using advanced analytics to solve a specific quality issue related to one of the company’s dairy products.
It also touches the topic of spreading awareness about the advantages of digital technologies in an industrial context.

Results:
An advanced analytics solution has been assessed and evaluated as capable of solving the quality problem. It will be deployed in a real production line and validated in the last quarter of 2017.
Many deliverables about machine learning theory, the digital technologies for the industry, the modelling of a digital factory has been produced and would remain a reference for spreading digital awareness within Nestlé.

Summary:
With 40 factories around the world, the company proposed a challenging environment for spare parts inventory management. They are all working independently with limited number of shipments between them. Our goal will be to create a model able to support a pooling inventory strategy, to compute level of stock needed and also to manage quantities repartition on the network. The target of this new model will be spare parts with very little demand, a high value, and common to several plants.

Results:
After literature search and analysis, we decided to create a new model based on categorization principle. In the huge number of SKU, a lot of articles have common points: same demand, same range of price, equivalent lead time, etc. To compute the reorder point, which represents the minimum stock level needed before ordering new items, the algorithm will allocate one coefficient for each parameter. Parameters are the item price, the lead time, the factory level in the hierarchy, etc. At the end, all these coefficients are multiplied to obtain the needed reorder point.