JMJulio Montecinos

Julio Montecinos

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Julio Montecinos has a total of 25 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Scalable Transit Delay Prediction at City Scale: A Systematic Approach with Multi-Resolution Feature Engineering and Deep Learning

    by Emna Boudabbous, Mohamed Karaa, Lokman Sboui, Julio Montecinos, Omar Alam

    Urban bus transit agencies need reliable, network-wide delay predictions to provide accurate arrival information to passengers and support real-time operational control. Accurate predictions help passengers plan their trips, reduce waiting time, and allow operations staff to adjust headways, dispatch extra vehicles, and manage disruptions. Although real-time feeds such as GTFS-Realtime (GTFS-RT) are now widely available, most existing delay prediction systems handle only a few routes, depend on hand-crafted features, and offer little guidance on how to design a scalable, reusable architecture. We present a city-scale prediction pipeline that combines multi-resolution feature engineering, dimensionality reduction, and deep learning. The framework generates 1,683 spatiotemporal features by exploring 23 aggregation combinations over H3 cells, routes, segments, and temporal patterns, and compresses them into 83 components using Adaptive PCA while preserving 95% of the variance. To avoid the "giant cluster" problem that occurs when dense urban areas fall into a single H3 region, we introduce a hybrid H3+topology clustering method that yields 12 balanced route clusters (coefficient of variation 0.608) and enables efficient distributed training. We compare five model architectures on six months of bus operations from the Société de transport de Montréal (STM) network in Montréal. A global LSTM with cluster-aware features achieves the best trade-off between accuracy and efficiency, outperforming transformer models by 18 to 52% while using 275 times fewer parameters. We also report multi-level evaluation at the elementary segment, segment, and trip level with walk-forward validation and latency analysis, showing that the proposed pipeline is suitable for real-time, city-scale deployment and can be reused for other networks with limited adaptation.

  • Navigating control and performance in emergency NGOs: the pivotal role of accountability

    by Muriel Dufour, Paulina Arroyo, Julio Montecinos

    Purpose: This study aims to investigate the pivotal role of accountability in shaping control mechanisms within non-governmental organizations (NGOs) engaged in disaster relief, examining how it influences control practices and operational capacity in resource management. Design/methodology/approach: We conducted a comparative case study based on 65 interviews with NGO experts in emergency crises and more than 300 documents. Findings: Our findings highlight accountability as a driving factor that shapes the formality and coerciveness of control mechanisms, often in response to donor demands. Emergency NGOs apply various forms of formal and informal control, with financial transparency being crucial. While intrinsic NGO characteristics alone could not explain differing control configurations, we found that accountability affects operational capacity within each NGO type through its influence on control formalization and coerciveness. Originality/value: Our findings challenge the belief that NGOs' intrinsic characteristics determine organizational control. Instead, accountability shapes control formalization. Our study offers practical insights for NGOs on how control choices affect operational capacity, examining multiple emergency NGOs in similar contexts rather than focusing on individual organizations.


  • Open Access

    Integrated Strategic and Tactical Design of Multi-Echelon City Distribution Systems with Vehicles Synchronization: Case of the Greater Montréal Area

    by Viviane Agnimo, Mustapha Ouhimmou, Marc Paquet, Julio Montecinos

    In this paper, we address a novel two-echelon City Logistics networks design problem that incorporates the new constraint of both echelons’ vehicles synchronization into the integrated strategic and tactical City Logistics networks design problem well known as location routing problem (LRP). The LRP combines both Facility Location Problem (FLP) at strategical decisions level and Vehicle Routing Problem (VRP) at tactical decisions level in City Logistics systems designing. We denote this new proposed problem Two-Echelon Capacitated Location Routing Problem with Vehicles Synchronization (2E-CLRPVS) and formulate a large-scale mixed integer programming model for it. Since based on literature LRPs to which belongs our problem are class of NP-hard problems, to enable solving the problem through exact solution method within reasonable computational time, we resort to a simplification technique relying on the formulation of a Two-Echelon Location-Allocation Problem (2E-LAP) from the 2E-CLRPVS and the fixing the binary variables of the 2E-LAP to solve the 2E-CLRPVS. Computational experiments are lead using the commercial software AMPL (A Mathematical Programming Language) on the case study of the greater Montréal area. To assess the validity of the proposed solution method, we perform comparison between the results obtained through this method and the optimal solution obtained by solving directly the 2E-CLRPVS in one step through AMPL. The computational results prove that the proposed solution method significantly reduces solving time, allows solving large instances and can find high-quality solutions. Extensive sensitivity analyzes have been conducted, to evaluate the impact of key parameters of the problem namely satellite, 1st echelon and 2nd echelon vehicle capacities on some important design performance indicators such as the total cost of the best objective, 1st echelon and 2nd echelon vehicles loading rates, vehicles waiting times at satellites. These analyzes can provide useful insights to practitioners in planning cost efficient and sustainable city distribution networks.

  • Top Cited Papers

    Gender diversity on board committees and ESG disclosure: evidence from Canada

    by Hanen Khemakhem, Paulina Arroyo, Julio Montecinos


  • Vehicle routing problem for reverse logistics of End-of-Life Vehicles (ELVs)

    by Amin Chaabane, Julio Montecinos, Mustapha Ouhimmou, Ahmed Khabou


  • Developing the seaport container terminal layout to enhance efficiency of the intermodal transportation system and port operations – case of the Port of Montreal

    by Tareq Abu Aisha, Mustapha Ouhimmou, Marc Paquet, Julio Montecinos

    The intermodal container terminal is considered the hearth of intermodal transportation systems, where different transportation modes meet to exchange containers. Using intermodal transportation efficiently reduces the handling procedures and relieves congestion at every transfer site. The efficient container terminal operations play a significant role in enhancing container flows at the container terminal. One way to enhance port operations efficiency and facilitate smooth container flow is to improve the intermodal container terminal layout. An efficient container terminal layout can be achieved through a reduction in the distance between the quay and rail tracks. We propose a novel layout to reduce the cost and accelerate container flow, thus improving the effectiveness of the container terminal. A mathematical model is developed to evaluate both layouts' performance considering congestion between nodes throughout the network. The Montreal Port is used as a case study to validate the proposed layout and the experiments conducted. It has been found that the proposed layout resulted in significant reductions in cost and time. Comparative results between the proposed layout and that of the Montreal Port demonstrates a reduction of 10% and 8% respectively to the total cost and time of transport containers from the port to their final destination.


  • Integrated forest harvest planning and road-building model with consideration of economies of scale

    by Azadeh Mobtaker, Julio Montecinos, Mustapha Ouhimmou, Mikael Rönnqvist

    We consider the problem of tactical forest management over a 5-year horizon with yearly periods. The main decisions made consider which harvest areas to cut in each period, the flow of timber from an area to each wood-processing mill to satisfy its annual demand, and which roads to build to access a harvest area not connected to the existing road network. The goal is to minimize the total transportation and road-building costs subject to budget limitations. To explore the benefits of economies of scale (EOS) in road construction, we incorporated this notion in the proposed model. Then, the efficiency of the obtained solution is compared with the model without EOS. The proposed model is a mixed-integer linear program, including several timber assortments and multiple periods. We validated the model for a realistic case in the context of the province of Quebec. The results demonstrate that consideration of EOS significantly reduces the total cost by about 5.3%. In the EOS solution, the road segments that are built every year are very concentrated in specific parts of the region, allowing a road-building company to take advantage of EOS. Moreover, this solution provides a much more efficient timber transportation plan.


  • Transport carriers’ cooperation on the last-mile delivery in urban areas

    by Julio Montecinos, Mustapha Ouhimmou, Satyaveer Chauhan, Marc Paquet, Ali Gharbi


  • Minimizing spatial dispersion of forest harvest areas using spectral clustering and set covering modelling

    by Azadeh Mobtaker, Julio Montecinos, Mustapha Ouhimmou, Mikael Rönnqvist, Marc Paquet

    In recent years, inclusion of spatial considerations in forest operations management has attracted great attention of both researchers and practitioners. In the province of Quebec, Canada, forest product companies subcontract harvesting operations to contractors. One of the challenges faced by the harvesting teams relates to moving the harvesting machineries between harvest areas, which is usually very costly and time consuming. To facilitate these operations, we propose a planning support tool to group the harvest areas in a way that the spatial dispersion of the clusters is reduced. We employed the spectral clustering algorithm to generate many alternative clusters of a set of harvest areas based on their transportation distance from one another and their available timber volume. Afterwards, a set covering model is developed to choose the clusters corresponding to the least spatial dispersion and approximately equal volume of timber. The approach is tested in a real case in Quebec and the results of two versions of set covering model were compared and analyzed.