List of topics available for students

  • AI and outsourcing: New trends and opportunities
    Operations Management / Project Scheduling / Machien/Deep Learning
    In the modern world where businesses face the challenge of the rapid growth of competitors, only those who lead their business more effectively manage to survive and reach success. Outsourcing is one of the techniques that can help the business to reach advantages above opponents. Outsourcing has become more popular since international businesses have started to differentiate between, on the one hand, their core business processes in which they have the necessary expertise and which often provide them with the maximal profit and, on the other hand, the other auxiliary processes which usually have a rather supporting role for the sustainability of their business. There are several decisions to be made, and thus I propose three directions for research: 1. Some resources may be booked multiple times. When the handling cost is substantially smaller than the total renting cost, and when there are no physical or operational limitations, the project manager may decide to book the external resource more than once for multiple periods. Such a decision would be probable especially if a single booking leads to very long resource idle times. Allowing infinite bookings, the problem structure becomes very similar to that of the resource renting problem. 2. Installing a resource may not be possible if certain other resources are already installed, and as such their renting periods should not overlap. A pair of resources that cannot be rented for overlapping periods is referred to as a conflict pair. Conflict pairs are given in form of a set. Each pair of resources in such a set must not be present on-site at the same time (i.e., must not be rented for the same period). 3. Some tasks may only require one of several alternative resources. For example, a lifting task could be executed by either a crane or a lift. Developing a website is another example, for which we need either a PHP developer or a Ruby on Rails developer. Logical resource requirement expressions, which can be simple or rather complex, determine the resource requirements for activities. An example of a complex case is when a certain cutting activity requires either a combination of a semi-automated cutting machine and a technician or a single fully automated cutting machine. How can we use AI-related approaches to deal with the above decision-making problems?
  • Reaction to disruptions in project planning: Learning-based approaches
    Operations Management / Project Scheduling / Machien/Deep Learning
    No matter how well we plan a project, disruptions are inevitable in a highly uncertain environment. Project managers/planners take different measures to resolve disruptions in project plans. An action taken to resolve the disruption is called the reaction or, in some fields, the recourse. A new trend has been emerged to use machine learning to obtain near-optimal reaction policies. Can we train models that based on existing exact models help us build fast reaction policies?
  • Machine/Deep learning approaches in project resource planning under uncertainty. (Topic 2)
    Operations Management / Project Scheduling / Machien/Deep Learning
    This research focuses on deep learning approaches to facilitate decision-making in a project with a high level of uncertainty. The goal is to find the answer to the question “How do we allocate resources?” If uncertain events interfere with the availability of resources in a project. Can we train a machine learning approach (say, a neural network) using past data that can help us make the best possible robust decision in terms of allocation of resources? The answer to the question is the goal of the thesis.
  • Machine/Deep learning approaches in project activity planning under uncertainty. (Topic 1)
    Operations Management / Project Scheduling / Machien/Deep Learning
    This research focuses on deep learning approaches to facilitate decision-making in a project with a high level of uncertainty. The goal is to find the answer to the question “When do we start an activity?” if uncertain events interfere with the project plan. An example is when not predicted COVID-related restrictive measures prolong some activities in a project so much that the start time of many other related activities (usually the successors) are heavily impacted. Can we train a machine learning approach (say, a neural network) using past data that can help us make the best possible robust decision? The answer to the question is the goal of the thesis.
  • The organ exchange problem: how to optimize the exchanges of organs between pairs of patient-doner?
    Operations Management / Optimization / Linear Programming

    Research Paper

  • To optimize the movement of chambers of a lock system
    Operations Management / Optimization / Linear Programming

    Research Paper

  • The role of AI in sharing economy
    Operations Management / AI

    Literature Review Study

This list is updated regularly. Visit this page again and you may find new topics potentially

You want to choose your own topic?


Contact me via email and we will discuss your ideas.

Designed by HashtBit.