http://hdl.handle.net/10468/1039 2017-10-30T00:06:35Z 2017-10-30T00:06:35Z Curran, Dara van der Krogt, Roman Little, James Wilson, Nic http://hdl.handle.net/10468/1405 2014-02-25T03:00:31Z 2013-11-01T00:00:00Z

Compilation of maintenance schedules based on their key performance indicators for fast iterative interaction Curran, Dara; van der Krogt, Roman; Little, James; Wilson, Nic Finding the optimal solution for a scheduling problem is hard, both from a computational perspective and because it is frequently difficult to articulate what the user wants. Often there are a range of possible key performance indicators (such as makespan, resource utilisation, and priority of tasks), and thus here can be many objectives that we want to optimise. However, it will typically be hard for the user to state numerical trade-offs between these objectives. Instead, it can be helpful if the user can explore the solution space themselves, to find which compromises between objectives they prefer. This paper demonstrates the use of Multi-valued Decision Diagrams in consideration of scheduling a real maintenance problem, namely the scheduling of Irish Navy dockyard maintenance. We show how candidate schedules can be compiled into MDDs, based on their associated Key Performance Indicators (KPIs). This representation allows the possible values of KPIs to be restricted by the user, and achievable values of other KPIs can be quickly determined, thus enabling fast iterative interaction with the user in order to achieve a satisfactory balance between the KPIs. We experimentally compare the performance of the MDD with that of a database, showing that the MDD can be considerably faster.

2013-11-01T00:00:00Z Manna, Carlo Fay, Damien Brown, Kenneth N. Wilson, Nic http://hdl.handle.net/10468/1393 2014-02-19T03:00:14Z 2013-11-01T00:00:00Z

Learning occupancy in single person offices with mixtures of multi-lag Markov chains Manna, Carlo; Fay, Damien; Brown, Kenneth N.; Wilson, Nic The problem of real-time occupancy forecastingfor single person offices is critical for energy efficient buildings which use predictive control techniques. Due to the highly uncertain nature of occupancy dynamics, the modeling and prediction of occupancy is a challenging problem. This paper proposes an algorithm for learning and predicting single occupant presence in office buildings, by considering the occupant behaviour as an ensemble of multiple Markov models at different time lags. This model has been tested using real occupancy data collected from PIR sensors installed in three different buildings and compared with state of the art methods, reducing the error rate by on average 5% over the best comparator method.

2013-11-01T00:00:00Z Trabelsi, Walid Wilson, Nic Bridge, Derek G. http://hdl.handle.net/10468/1404 2014-02-25T03:00:14Z 2013-11-01T00:00:00Z

Comparative preferences induction methods for conversational recommenders Trabelsi, Walid; Wilson, Nic; Bridge, Derek G. Perny, Patrice; Pirlot, Marc; Tsoukiàs, Alexis In an era of overwhelming choices, recommender systems aim at recommending the most suitable items to the user. Preference handling is one of the core issues in the design of recommender systems and so it is important for them to catch and model the user’s preferences as accurately as possible. In previous work, comparative preferences-based patterns were developed to handle preferences deduced by the system. These patterns assume there are only two values for each feature. However, real-world features can be multi-valued. In this paper, we develop preference induction methods which aim at capturing several preference nuances from the user feedback when features have more than two values. We prove the efficiency of the proposed methods through an experimental study.

2013-11-01T00:00:00Z Marinescu, Radu Razak, Abdul Wilson, Nic http://hdl.handle.net/10468/1402 2014-02-20T03:00:09Z 2013-09-01T00:00:00Z

Multi-objective constraint optimization with tradeoffs Marinescu, Radu; Razak, Abdul; Wilson, Nic Schulte, Christian In this paper, we consider the extension of multi-objective constraint optimization algorithms to the case where there are additional tradeoffs, reducing the number of optimal solutions. We focus especially on branch-and-bound algorithms which use a mini-buckets algorithm for generating the upper bound at each node (in the context of maximizing values of objectives). Since the main bottleneck of these algorithms is the very large size of the guiding upper bound sets we introduce efficient methods for reducing these sets, yet still maintaining the upper bound property. We also propose much faster dominance checks with respect to the preference relation induced by the tradeoffs. Furthermore, we show that our tradeoffs approach which is based on a preference inference technique can also be given an alternative semantics based on the well known Multi-Attribute Utility Theory. Our comprehensive experimental results on common multi-objective constraint optimization benchmarks demonstrate that the proposed enhancements allow the algorithms to scale up to much larger problems than before.

2013-09-01T00:00:00Z