M.Sc. Data Analytics and Decision Science

• Network Planning: Planning transportation in an effective manner, deciding about locations, company-wide and within logistic nodes, can save a company millions of dollars. These decisions are inherently complex, with interdependent decisions under involved constraints. Leveraging the full potential requires modelling and solving the decision situations using the latest mathematical optimization techniques.
• Replenishment: How many items of a specific article are needed at a particular store or production location? What may sound like a simple question is a difficult challenge in practice: Retailers and industrial production processes depend on thousands of products at multiple locations which are sourced from a complex supply chain network. In practice, a large number of additional constraints such as varying lot-sizes, delivery or production lead-times, etc. complicate the replenishment process further.  Optimal replenishment and procurement decisions avoid stockout situations while lowering capital cost or waste of perishable items and thus increasing the overall profit.
• Dynamic Pricing: One of the core competences of any retailer is to set the best price of any product at any time. However, as many retailers stock tens of thousands of products at hundreds of sales locations, finding the right price for each of them every day is a Herculean task, in particular, as customer behavior shifts and product sales are influenced by a myriad of factors. Building robust systems setting the right price automatically according to a specific business strategy is paramount to the success of the company

The next industrial revolution termed “Industry 4.0” has just begun: Apart from highly visible applications such as self-driving cars, many industrial processes can benefit significantly from data-driven decisions. To exploit the full potential it is not sufficient (or even counter productive) to massively collect data, but to collect exactly the right data in order to feed them into predictive models and base decisions on them using prescriptive modeling.

• Determining production sequences on machines and respective lot sizes are everyday tasks, however, these decisions are strongly combinatorial in nature. Unlike algorithmic approaches, humans are generally not able to take the multitude of correlated influencing factors into account in order to make optimal decisions.
• Although today’s production machines and industrial robots are very sophisticated, operating them often requires decades of specialized training to gain the relevant experience to operate these machines optimally. Using data-based methods, assistance systems and self-regulating machines capture a significant part of the expertise and aim to make the production process more efficient.
• By combining measurements of the item currently being produced, other data sources and external information such as the order book, production lines can be optimized dynamically to improve the pipeline: Which item can be produced now with the materials at hand? Does the quality of the individual product match the requirements from the order, are further production steps required or should the production line be changed dynamically to fulfil a different order first?
• All industrial production units have to undergo regular maintenance. Planning the best maintenance period is critical for most production lines to optimize the overall yield. Predictive maintenance allows to detect otherwise unnoticed signs of a forthcoming machine breakdown or tool wear. Based on this information, and also depending on the production schedule, the availability of spare parts, maintenance personnel, open orders etc. one builds prescriptive models to optimally schedule maintenance operations (prescriptive maintenance).

• Airlines were among the first to adopt operations research technology in their strategic, tactical and operations planning. Typical airline planning tasks arise in network planning, aircraft routing, maintenance scheduling, crew assignment, disruption management etc. Only mathematical optimization techniques can capture the combinatorial variety of these tasks which appear in similar form when operating trains or other kinds of public transport.
• Autonomous driving, in particular based on electric cars, requires to re-consider known planning problems in a new light. Locating charging or changing stations for cars,  coordinating or balancing the re-charging of cars in order to avoid peaks in electricity demand, etc. are only a few examples. Predictive and prescriptive analytics work hand in hand in these fields.
• Public transportation providers face planning tasks such as network layout, line planning, creation of time-tables, vehicle and staff planning as well as managing operational disruptions in real-time. Traditional engineering approaches typically use a hierachical approach to deal with these tasks, modern optimization techniques however enable companies to solve these challenges in an integrated, more holistic approach.
• Traffic management features such as advanced parking solutions, virtual street signs, sharing business models, etc. will be connected via integrated mobility platforms to keep cities free of traffic jams and unnecessary emissions.

• Pricing emissions: The causes and economic consequences of climate change are becoming increasingly important for investment decisions. In order to estimate the long run effects of emissions of greenhouse gases on the economy, complex climate-economy models need to be formulated, calibrated and solved. This task requires the development and application of modern quantitative methods for solving complex, high-dimensional, non-linear dynamic optimization problems with explicit treatment of uncertainty.
• Policy consulting: Decision-making regarding sustainable growth requires policy makers to take the intrinsically uncertain nature of future states of the economy and the earth system components into account. Facing this important problem, decision makers have to choose an appropriate portfolio of actions and policies to increase the welfare return to society. Thus, decision makers face the risk management problem of designing an energy and climate policy that is robust to a broad range of uncertain future conditions.

• Marketing (e.g. ad placement, SEO optimization, marketing campaign optimization, …)
• Health care (e.g. personalized medicine, automated diagnostics, …)
• Finance & Insurance (e.g. predictive underwriting, fraud detection, tariff optimisation, …)
• Politics & public sector
• Agriculture
• Sustainability & Smart Living
• Sports
• Education
• …

In fact, it is hard to come up with an industry or service that is not directly or indirectly impacted by data and algorithms driven decision making. As the need for decisions changes (more flexible, real-time, under uncertain and ever changing environments), a  decision making process rooted in mathematical optimization is unavoidable. Only predictive analytics (e.g. machine learning) can harness the potential of historic and present data; only prescriptive analytics (mathematical optimization and operations research) can capture the full range of options the decision maker faces.

The M.Sc. degree granted by RWTH Aachen University in  Germany will also enable you to pursue an academic career and continue studying towards a PhD in fields such as data analytics, machine learning & artificial intelligence, operations research and engineering.

Whichever way you want to follow after graduation: Our dedicated team of experts in the career and entrepreneurship centers will accompany you on that journey and help you decide how to best realize your ambitions, be it an exciting new job or starting your own business. We seek to place our graduates in global technology blue chips, hidden technology champions, leading technology consultancies and fast-growing technology ventures. The Entrepreneurship Center at RWTH Aachen University also has a long track record of supporting innovative startups by our graduates – between 70 to 100 every year.

This program is ideal for you if you want to develop as a professional and transform your career, provided:

• You have a STEM (Science, Technology, Engineering and Mathematics) background, at least one year of full-time work experience and want to deepen your knowledge in machine learning, artificial intelligence, operations research techniques (mathematical optimization, heuristic algorithms and simulation), and data-driven decision making.
• You are passionate about engineering and technology and want to enhance your skills to face tomorrow’s challenges in creating value from data using algorithms.
• You want to learn how to leverage machine learning, artificial intelligence, operations research, and data-driven decision making into profitable and sustainable business models that allow you to lead the technological transformation in your industry rather than just following it.
• You have a good knowledge and working experience with at least one high level programming language (e.g. Python, Java, C/C++), as well as some experience in developing software or contributing to a software project.