Luís Guimarães

Abstract
In this paper, we explore the value of considering simultaneous pickups and deliveries inmulti-product inventory-routing problems both with deterministic and uncertain demand. Wepropose a multi-commodity, develop an exact branch-and-cut algorithm with patching heuristicsto efficiently tackle this problem, and provide insightful analyses based on optimal plans. Thesimplicity of the proposed approach is an important aspect, as it facilitates its usage in practice,opposed to complicated stochastic or probabilistic methods. The computational experimentssuggest that in the deterministic demand setting, pickups are mainly used to balance initialinventories, achieving an average total cost reduction of 1.1%, while transshipping 2.4% oftotal demand. Under uncertain demand, pickups are used extensively, achieving cost savings of up to 6.5% in specific settings. Overall, our sensitivity analysis shows that high inventory costsand high degrees of demand uncertainty drive the usage of pickups, which, counter-intuitively, are not desirable in every case

Abstract
Distribution warehouses are a critical part of supply chains, representing a nonnegligible share of the operating costs. This is especially true for unautomated, labor-intensive warehouses, partially due to time-consuming activities such as picking up items or traveling. Inventory categorization techniques, as well as zone storage assignment policies, may help in improving operations, but may also be short-sighted. This work presents a three-step methodology that uses probabilistic simulation, optimization, and event-based simulation (SOS) to analyze and experiment with layout and storage assignment policies to improve the picking performance. In the first stage, picking performance is estimated under different storage assignment policies and zone configurations using a probabilistic model. In the second stage, a mixed integer optimization model defines the overall warehouse layout by selecting the configuration and storage assignment policy for each zone. Finally, the optimized layout solution is tested under demand uncertainty in the third, final simulation phase, through a discrete-event simulation model. The SOS methodology was validated with three months of operational data from a large retailer's warehouse, successfully illustrating how it may be successfully used for improving the performance of a distribution warehouse.

Abstract
The fast-moving consumer goods sector relies on economies of scale. However, its assortments have been overextended as a means of market share appropriation and top-line growth. This paper studies the se-lection of the optimal set of products for fast-moving consumer goods producers to offer, as there is no previous model for product line selection that satisfies the requirements of the sector. Our mixed -integer programming model combines a multi-category attraction model with a capacitated lot-sizing problem, shared setups and safety stock. The multi-category attraction model predicts how the demand for each product responds to changes within the assortment. The capacitated lot-sizing problem allows us to account for the indirect production costs associated with different assortments. As seasonality is prevalent in consumer goods sales, the production plan optimally weights the trade-off between stocking finished goods from a long run with performing shorter runs with additional setups. Finally, the safety stock extension addresses the effect of the demand uncertainty associated with each assortment. With the computational experiments, we assess the value of our approach using data based on a real case. Our findings suggest that the benefits of a tailored approach are at their highest in scenarios typical fast-moving consumer goods industry: when capacity is tight, demand exhibits seasonal patterns and high service levels are required. This also occurs when the firm has a strong competitive position and consumer price-sensitivity is low. By testing the approach in two real-world instances, we show that this decision should not be made based on the current myopic industry practices. Lastly, our approach obtains profits of up to 9.4% higher than the current state-of-the-art models for product line selection.

Abstract
When making long-term plans for their asset portfolios, decision-makers have to define a priori a maintenance budget that is to be shared among the several assets and managed throughout the planning period. During the planning period, the a priori budget is then allocated by managers to different operation and maintenance interventions ensuring the overall performance of the system. Because asset degradation is stochastic, a considerable amount of uncertainty is associated with this problem. Hence, to define a robust budget, it is essential to account for several degradation scenarios pertaining to the individual condition of each asset. This paper presents a novel mathematical formulation to tackle this problem in a heterogeneous multiasset portfolio. The proposed mathematical model was formulated as a mixed-integer programming two-stage stochastic optimization model with mean-variance constraints to minimize the number of scenarios with an insufficient budget. A Gamma process was used to model the condition of each individual asset while taking into consideration different technological features and operating conditions. We compared the solutions obtained with our model to alternative practices in a set of generated instances covering different types of multi-asset portfolios. This comparison allowed us to explore the value of modeling uncertainty and how it affects the generated solutions. The proposed approach led to gains in performance of up to 50% depending on the level of uncertainty. Furthermore, the model was validated using real-world data from a utility company working with portfolios of power transformers. The results obtained showed that the company could reduce costs by as much as 40%. Further conclusions showed that the cost-saving potential was higher in asset portfolios in worse condition and that defining a priori operation and maintenance interventions led to worse results. Finally, the results showcased how different decision-maker risk-levels affect the value of taking uncertainty into account.

Abstract
Electrical utilities apply condition monitoring on power transformers (PTs) to prevent unplanned outages and detect incipient faults. This monitoring is often done using dissolved gas analysis (DGA) coupled with engineering methods to interpret the data, however the obtained results lack accuracy and reproducibility. In order to improve accuracy, various advanced analytical methods have been proposed in the literature. Nonetheless, these methods are often hard to interpret by the decision-maker and require a substantial amount of failure records to be trained. In the context of the PTs, failure data quality is recurrently questionable, and failure records are scarce when compared to nonfailure records. This work tackles these challenges by proposing a novel unsupervised methodology for diagnosing PT condition. Differently from the supervised approaches in the literature, our method does not require the labeling of DGA records and incorporates a visual representation of the results in a 2D scatter plot to assist in interpretation. A modified clustering technique is used to classify the condition of different PTs using historical DGA data. Finally, well-known engineering methods are applied to interpret each of the obtained clusters. The approach was validated using data from two different real-world data sets provided by a generation company and a distribution system operator. The results highlight the advantages of the proposed approach and outperformed engineering methods (from IEC and IEEE standards) and companies legacy method. The approach was also validated on the public IEC TC10 database, showing the capability to achieve comparable accuracy with supervised learning methods from the literature. As a result of the methodology performance, both companies are currently using it in their daily DGA diagnosis.