Multiple Object Tracking (MOT) is a rapidly developing research field that targets precise and reliable tracking of objects. Unfortunately, most available MOT datasets typically contain short video clips only, disregarding the indispensable requirement for adequately capturing substantial long-term variations in real-world scenarios. Long-term MOT poses unique challenges due to changes in both the objects and the environment, which remain relatively unexplored. To fill the gap, we propose a time-lapse image dataset inspired by the growth monitoring of strawberries, dubbed The Growing Strawberries Dataset (GSD). The data was captured hourly by six cameras, covering a span of 16 months in 2021 and 2022. During this time, it encompassed a total of 24 plants in two separate greenhouses. The changes in appearance, weight, and position during the ripening process, along with variations in the illumination during data collection, distinguish the task from previous MOT research. These practical issues resulted in a drastic performance downgrade in the track identification and association tasks of state-of-the-art MOT algorithms. We believe The Growing Strawberries will provide a platform for evaluating such long-term MOT tasks and inspire future research. The dataset is available at https://doi.org/10.4121/e3b31ece-cc88-4638-be10-8ccdd4c5f2f7.v1.

@en

The project “Real-time data-driven maintenance logistics” was initiated with the purpose of bringing innovations in data-driven decision making to maintenance logistics, by bringing problem owners in the form of three innovative companies together with researchers at two leading knowledge institutions. This paper reviews innovations in three related areas: How the innovations were inspired by practice, how they materialized, and how the results impact practice.

@en

Train routing is sensitive to delays that occur in the network. When a train is delayed, it is imperative that a new plan be found quickly, or else other trains may need to be stopped to ensure safety, potentially causing cascading delays. In this paper, we consider this class of multi-agent planning problems, which we call Multi-Agent Execution Delay Replanning. We show that these can be solved by reducing the problem to an any-start-time safe interval path planning problem. When an agent has an any-start-time plan, it can react to a delay by simply looking up the precomputed plan for the delayed start time. We identify crucial real-world problem characteristics like the agent's speed, size, and safety envelope, and extend the any-start-time planning to account for them. Experimental results on real-world train networks show that any-start-time plans are compact and can be computed in reasonable time while enabling agents to instantly recover a safe plan.

@en

In reinforcement learning (RL), different reward functions can define the same optimal policy but result in drastically different learning performance. For some, the agent gets stuck with a suboptimal behavior, and for others, it solves the task efficiently. Choosing a good reward function is hence an extremely important yet challenging problem. In this paper, we explore an alternative approach for using rewards for learning. We introduce max-reward RL, where an agent optimizes the maximum rather than the cumulative reward. Unlike earlier works, our approach works for deterministic and stochastic environments and can be easily combined with state-of-the-art RL algorithms. In the experiments, we study the performance of max-reward RL algorithms in two goal-reaching environments from Gymnasium-Robotics and demonstrate its benefits over standard RL. The code is available at https://github.com/veviurko/To-the-Max.

@en

When optimizing problems with uncertain parameter values in a linear objective, decision-focused learning enables end-to-end learning of these values. We are interested in a stochastic scheduling problem, in which processing times are uncertain, which brings uncertain values in the constraints, and thus repair of an initial schedule may be needed. Historical realizations of the stochastic processing times are available. We show how existing decision-focused learning techniques based on stochastic smoothing can be adapted to this scheduling problem. We include an extensive experimental evaluation to investigate in which situations decision-focused learning outperforms the state of the art, i.e., scenario-based stochastic optimization.

@en

People want to rely on optimization algorithms for complex decisions but verifying the optimality of the solutions can then become a valid concern, particularly for critical decisions taken by non-experts in optimization. One example is the shortest-path problem on a network, occurring in many contexts from transportation to logistics to telecommunications. While the standard shortest-path problem is both solvable in polynomial time and certifiable by duality, introducing side constraints makes solving and certifying the solutions much harder. We propose a proof system for constrained shortest-path problems, which gives a set of logical rules to derive new facts about feasible solutions. The key trait of the proposed proof system is that it specifically includes high-level graph concepts within its reasoning steps (such as connectivity or path structure), in contrast to using linear combinations of model constraints. Using our proof system, we can provide a step-by-step, human-auditable explanation showing that the path given by an external solver cannot be improved. Additionally, to maximize the advantages of this setup, we propose a proof search procedure that specifically aims to find small proofs of this form using a procedure similar to A* search. We evaluate our proof system on constrained shortest path instances generated from real-world road networks and experimentally show that we may indeed derive more interpretable proofs compared to an integer programming approach, in some cases leading to much smaller proofs.

@en

Global soft fruit supply chains rely on trustworthy descriptions of product quality. However, crucial criteria such as sweetness and firmness cannot be accurately established without destroying the fruit. Since traditional alternatives are subjective assessments by human experts, it is desirable to obtain quality estimations in a consistent and non-destructive manner. The majority of research on fruit quality measurements analyzed fruits in the lab with uniform data collection. However, it is laborious and expensive to scale up to the level of the whole yield. The “harvest-first, analysis-second” method also comes too late to decide to adjust harvesting schedules. In this research, we validated our hypothesis of using in-field data acquirable via commodity hardware to obtain acceptable accuracies. The primary instance that the research concerns is the sugariness of strawberries, described by the juice’s total soluble solid (TSS) content (unit: °Brix or Brix). We benchmarked the accuracy of strawberry Brix prediction using convolutional neural networks (CNN), variational autoencoders (VAE), principal component analysis (PCA), kernelized ridge regression (KRR), support vector regression (SVR), and multilayer perceptron (MLP), based on fusions of image data, environmental records, and plant load information, etc. Our results suggest that: (i) models trained by environment and plant load data can perform reliable prediction of aggregated Brix values, with the lowest RMSE at 0.59; (ii) using image data can further supplement the Brix predictions of individual fruits from (i), from 1.27 to as low up to 1.10, but they by themselves are not sufficiently reliable.@en

Surrogate algorithms such as Bayesian optimisation are especially designed for black-box optimisation problems with expensive objectives, such as hyperparameter tuning or simulation-based optimisation. In the literature, these algorithms are usually evaluated with synthetic benchmarks which are well established but have no expensive objective, and only on one or two real-life applications which vary wildly between papers. There is a clear lack of standardisation when it comes to benchmarking surrogate algorithms on real-life, expensive, black-box objective functions. This makes it very difficult to draw conclusions on the effect of algorithmic contributions and to give substantial advice on which method to use when. A new benchmark library, EXPObench, provides first steps towards such a standardisation. The library is used to provide an extensive comparison of six different surrogate algorithms on four expensive optimisation problems from different real-life applications. This has led to new insights regarding the relative importance of exploration, the evaluation time of the objective, and the used model. We also provide rules of thumb for which surrogate algorithm to use in which situation. A further contribution is that we make the algorithms and benchmark problem instances publicly available, contributing to more uniform analysis of surrogate algorithms. Most importantly, we include the results of the six algorithms on all evaluated problem instances. This unique new dataset lowers the bar for researching new methods as the number of expensive evaluations required for comparison and for the creation of new surrogate models is significantly reduced.

@en

Global optimization of decision trees has shown to be promising in terms of accuracy, size, and consequently human comprehensibility. However, many of the methods used rely on general-purpose solvers for which scalability remains an issue. Dynamic programming methods have been shown to scale much better because they exploit the tree structure by solving subtrees as independent subproblems. However, this only works when an objective can be optimized separately for subtrees. We explore this relationship in detail and show necessary and sufficient conditions for such separability and generalize previous dynamic programming approaches into a framework that can optimize any combination of separable objectives and constraints. Experiments on five application domains show the general applicability of this framework, while outperforming the scalability of general-purpose solvers by a large margin.@en

The Train Unit Shunting Problem concerns the parking of trains outside their scheduled use on so-called shunting yards. This is an NP-hard problem, and the current algorithm used by the Netherlands Railways cannot detect whether an instance is infeasible. So, infeasible instances can cause needlessly long computation times. Therefore, this paper fills the gap by providing novel approaches to determine the feasibility. For this, the Pebble Motion problem is considered which moves pebbles from their starting node to their goal node in the graph, such that no two pebbles occupy a node at the same time. A variant of the Pebble Motion problem is proposed to model the Train Unit Shunting Problem, where train units are represented by pebbles and the arrival and departure of train unit combinations are also included. This paper specifically looks at dead-end track shunting yards, as they can be abstractly represented by trees, such that trains arrive and depart at the root node. Furthermore, trains cannot be reallocated between arrival and departure in the tree, since reallocation in practice is a very costly process as moves need to be performed by a small set of drivers. The conditions for realizing the departure order of trains are studied, and an efficient method to (partially) determine the feasibility of problem instances is given, which can find the minimal number of tracks required to park the trains. Furthermore, a special case with tracks of length two is shown to be polynomially solvable, while another subset of problem instances with tracks of length six or more is demonstrated to be NP-complete.

@en

We study a highly complex scheduling problem that requires the generation and optimization of production schedules for a multi-product biomanufacturing system with continuous and batch processes. There are two main objectives here; makespan and lateness, which are combined into a cost function that is a weighted sum. An additional complexity comes from long horizons considered (up to a full year), yielding problem instances with more than 200 jobs, each consisting of multiple tasks that must be executed in the factory. We investigate whether a rolling-horizon principle is more efficient than a global strategy. We evaluate how cost function weights for makespan and lateness should be set in a rolling-horizon approach where deadlines are used for subproblem definition. We show that the rolling-horizon strategy outperforms a global search, evaluated on problem instances of a real biomanufacturing system, and we show that this result generalizes to problem instances of a synthetic factory.

@en

Predict and optimize is an increasingly popular decision-making paradigm that employs machine learning to predict unknown parameters of optimization problems. Instead of minimizing the prediction error of the parameters, it trains predictive models using task performance as a loss function. In the convex optimization domain, predict and optimize has seen significant progress due to recently developed methods for differentiating optimization problem solutions over the problem parameters. This paper identifies a yet unnoticed drawback of this approach -- the zero-gradient problem -- and introduces a method to solve it. The suggested method is based on the mathematical properties of differential optimization and is verified using two real-world benchmarks. @en

The integration of Distributed Energy Resources (DERs) in distribution networks comes with challenges, like power quality concerns, but also opens up new opportunities, e.g., DERs can offer competitive energy prices for final users by leveraging time arbitrage. A suitable method to fully exploit such opportunities is to compute the optimal DER schedule, either with a full three-phase network model or a more computationally efficient single-line equivalent. This paper presents under which conditions a single-line equivalent can and cannot be used to properly represent a modern and unbalanced power distribution network able to dispatch high levels of DER integration optimally. Results show that single-line equivalents might be helpful when the problem objective function limits counterflows, for example, when minimizing active power losses. Moreover, single-line equivalents might be helpful for low levels of DER integration. However, enabling single-line equivalents results in a lower hosting capacity for high levels of DER integration.

@en

Gene-pool Optimal Mixing Evolutionary Algorithms (GOMEAs) have been shown to achieve state-of-the-art results on various types of optimization problems with various types of problem variables. Recently, a GOMEA for permutation spaces was introduced by leveraging the random keys encoding, obtaining promising first results on permutation flow shop instances. A key cited strength of GOMEAs is linkage learning, i.e., the ability to determine and leverage, during optimization, key dependencies between problem variables. However, the added value of linkage learning was not tested in depth for permutation GOMEA. Here, we introduce a new version of permutation GOMEA, called qGOMEA, that works directly in permutation space, removing the redundancy of using random keys. We additionally consider various linkage information sources, including random noise, in both GOMEA variants, and compare performance with various classic genetic algorithms on a wider range of problems than considered before. We find that, although the benefits of linkage learning are clearly visible for various artificial benchmark problems, this is far less the case for various real-world inspired problems. Finally, we find that qGOMEA performs best, and is more applicable to a wider range of permutation problems.

@en

District heating system (DHS) optimization is becoming an increasingly important problem because of the unused potential in flexibility that could allow less energy being wasted and the integration of renewable energy. While new optimization methods are proposed every year to tackle this problem, the literature lacks a good way to benchmark newly proposed methods. To address this problem, we introduce GridPenguin, an open-source computational simulator for the physics of district heating networks. It provides flexibility in usage by providing building blocks with which the user can build any grid he wants. The detailed simulation of the physical world with a focus on the heat balance and average flow rate and temperature allows for fast and accurate simulation. By explaining the physical equations and computational model as well as the comparison to existing software, we lay a solid foundation for the performance of the simulator. We present GridPenguin as a metric to evaluate optimization methods as well as a tool for easy integration of advanced machine learning methods into DHS optimization. The source code of our project can be found on https://github.com/ftbv/grid-penguin.@en

Many electric vehicles (EVs) are using today’s distribution grids, and their flexibility can be highly beneficial for the grid operators. This flexibility can be best exploited by DC power networks, as they allow charging and discharging without extra power electronics and transformation losses. From the grid control perspective, algorithms for planning EV charging are necessary. This paper studies the problem of EV charging planning under limited grid capacity and extends it to the partially observable case. We demonstrate how limited information about the EV locations in a grid may disrupt the operation planning in DC grids with tight constraints. We introduce two methods to change the grid topology such that partial observability of the EV locations is resolved. The suggested models are evaluated on the IEEE 16 bus system and multiple randomly generated grids with varying capacities. The experiments show that these methods efficiently solve the partially observable EV charging planning problem and offer a trade-off between computational time and performance. @en

Wind farms suffer from so-called wake effects: when turbines are located in the wind shadows of other turbines, their power output is substantially reduced. These losses can be partially mitigated via actively changing the yaw from the individually optimal direction. Most existing wake control techniques have two major limitations: they use simplified wake models to optimize the control strategy, and they assume that the atmospheric conditions remain stable. In this paper, we address these limitations by applying reinforcement learning (RL). RL forgoes the wake model entirely and learns an optimal control strategy based on the observed atmospheric conditions and a reward signal, in this case the power output of the farm. It also accounts for random transitions in the observations, such as turbulent fluctuations in the wind. To evaluate RL for active wake control, we provide a simulator based on the state-of-the-art FLORIS model in the OpenAI gym format. Next, we propose three different state-action representations of the active wake control problem and investigate their effect on the performance of RL-based wake control. Finally, we compare RL to a state-of-the-art wake control strategy based on FLORIS and show that RL is less sensitive to changes in unobservable data.

@en

Interpretable and fair machine learning models are required for many applications, such as credit assessment and in criminal justice. Decision trees offer this interpretability, especially when they are small. Optimal decision trees are of particular interest because they offer the best performance possible for a given size. However, state-of-the-art algorithms for fair and optimal decision trees have scalability issues, often requiring several hours to find such trees even for small datasets. Previous research has shown that dynamic programming (DP) performs well for optimizing decision trees because it can exploit the tree structure. However, adding a global fairness constraint to a DP approach is not straightforward, because the global constraint violates the condition that subproblems should be independent. We show how such a constraint can be incorporated by introducing upper and lower bounds on final fairness values for partial solutions of subproblems, which enables early comparison and pruning. Our results show that our model can find fair and optimal trees several orders of magnitude faster than previous methods, and now also for larger datasets that were previously beyond reach. Moreover, we show that with this substantial improvement our method can find the full Pareto front in the trade-off between accuracy and fairness.@en