Proposed Solution Approach Research Articles

Deep Learning-Driven Analysis of a Six-Bar Mechanism for Personalized Gait Rehabilitation

Abstract Recent advances in robotics and artificial intelligence have highlighted the potential for the integration of computational intelligence in enhancing the functionality and adaptability of robotic systems, particularly in rehabilitation. Designing robotic exoskeletons for the lower limb rehabilitation of post-stroke patients requires frequent adjustments to accommodate individual differences in leg anatomy. This complex engineering challenge necessitates a deep understanding of human physiology, robotics, and optimization to develop adaptive robotic systems and also to swiftly quantify the required adjustments and implement them for each patient. The conventional approaches, which mostly rely on heuristics and manual tuning, often struggle to achieve optimal results. This paper presents a novel method that integrates a genetic algorithm (GA) with a deep learning approach to generate a gait trajectory of the ankle joint from a six-bar linkage mechanism of fixed dimensions. Later, using the same approach, the inverse kinematics solution for this mechanism is also devised whereby, the set of the link dimensions of the six-bar linkage mechanism is obtained for the given gait trajectory of an individual to achieve customization. We simulated the kinematic behavior of the six-bar linkage mechanism within defined mechanical constraints and utilized the generated data for training a feedforward neural network and long short-term memory models. The proposed model, when trained, can produce accurate lengths for the desired gait trajectories in the sagittal plane and vice-versa which further validates our proposed approach for inverse kinematics solution.

Designing a Fuzzy Mathematical Model for a Two-Echelon Allocation-Routing Problem by Applying Route Conditions: A New Interactive Fuzzy Approach

In vehicle routing problems (VRP), the optimal allocation of transportation by considering factors such as route hardness, driver experience and vehicle worn-out has a significant effect on costs reduction and approaching real-world conditions. In this paper, a novel fuzzy mixed integer non-linear mathematical model to address the two-echelon allocation-routing problem under uncertainty is proposed by applying route and fleet conditions. The cost of allocating drivers to diverse vehicles is computed at the first echelon of the problem, considering factors such as vehicle type, vehicle wear-out, and driver experience. Additionally, different routes are defused with varying levels of hardness. The goal of the second echelon of the model is to improve reliability by defining the reliability of routes within each section. To solve the model, the Torabi and Hessini (TH), the Selimi and Ozkarahan (SO) methods, and a newly proposed approach (PIA) were utilized to transform the multi-objective model into a single-objective one. Numerical tests and performance indicators were used to validate the effectiveness of both the multi-objective mathematical model and the proposed solution method. The validation computation results indicate that the proposed solution approach outperforms both the TH and SO approaches.

Essential parameters needed for a U-Net-based segmentation of individual bones on planning CT images in the head and neck region using limited datasets for radiotherapy application.

Objective.The field of radiotherapy is highly marked by the lack of datasets even with the availability of public datasets. Our study uses a very limited dataset to provide insights on essential parameters needed to automatically and accurately segment individual bones on planning CT images of head and neck cancer patients.Approach.The study was conducted using 30 planning CT images of real patients acquired from 5 different cohorts. 15 cases from 4 cohorts were randomly selected as training and validation datasets while the remaining were used as test datasets. Four experimental sets were formulated to explore parameters such as background patch reduction, class-dependent augmentation and incorporation of a weight map on the loss function.Main results.Our best experimental scenario resulted in a mean Dice score of 0.93 ± 0.06 for other bones (skull, mandible, scapulae, clavicles, humeri and hyoid), 0.93 ± 0.02 for ribs and 0.88 ± 0.03 for vertebrae on 7 test cases from the same cohorts as the training datasets. We compared our proposed solution approach to a retrained nnU-Net and obtained comparable results for vertebral bones while outperforming in the correct identification of the left and right instances of ribs, scapulae, humeri and clavicles. Furthermore, we evaluated the generalization capability of our proposed model on a new cohort and the mean Dice score yielded 0.96 ± 0.10 for other bones, 0.95 ± 0.07 for ribs and 0.81 ± 0.19 for vertebrae on 8 test cases.Significance.With these insights, we are challenging the utilization of an automatic and accurate bone segmentation tool into the clinical routine of radiotherapy despite the limited training datasets.

Efficient modeling and simulation of gas separations applying Maxwell-Stefan approach and Ideal Adsorbed Solution Theory

We present an efficient way to incorporate the Maxwell-Stefan approach (Krishna, 1990) for modeling gas separation processes, i.e., adsorbent particles and membranes. Ideal Adsorbed Solution Theory (Myers and Prausnitz, 1965) is applied for the description of competitive equilibria. The proposed calculation method introduces analytical expressions for the partial derivatives of the loadings w.r.t. fluid-phase concentrations, developed by Rubiera Landa et al. (2013), in the calculation of the thermodynamic correction factors, as required by Maxwell-Stefan flux equations. These expressions are solved simultaneously with transient mass balances that describe the gas separation processes in a differential-algebraic equation formulation. Robustness and advantages of the proposed solution approach are demonstrated with several examples adapted from the literature.

Modeling and solving the fuel distribution problem with unloading precedence and loading sequence considerations

AbstractThis paper presents a real-world liquid fuel distribution problem involving a heterogeneous fleet of multi-compartment vehicles servicing a set of orders of different fuels. Two new features are introduced that affect significantly the performance of the fuel distribution process in terms of safety and efficiency: (i) loading trucks so that the payload remains balanced throughout each phase of a delivery route, and (ii) sequencing the requests for loading trucks at the depot loading facilities. A Mixed Integer Programming formulation is presented and an Adaptive Large Neighbourhood Search algorithm with various novel features is developed and benchmarked. A new loading model is formed and solved for allocating the ordered items (fuel) to vehicle compartments, as part of constructing/repairing delivery routes. The computational performance of the proposed solution approach has been tested on a series of benchmark problems. Moreover, a series of experiments were performed in order to assess the effect of the balanced loading constraints on the traveled distance. The results indicate that the effect of this type of constraint on the total traveled distance is kept at a reasonable level, reaching a maximum increase of 4.37%. The computational tools presented in this work may accommodate the dispatchers’ work in producing efficient and safe delivery routes while managing efficiently any potential bottleneck in the truck loading facilities.

Process optimization of microbial fermentation with parameter uncertainties via distributionally robust discrete control

There are some uncertain kinetic parameters in microbial fermentation system because of the unclear intracellular metabolic mechanisms. Considering the affection of these uncertain parameters on system performance, dynamic process optimization of the fermentation system can be modeled as a distributionally robust discrete control problem under moment uncertainty, which aims to maximize the mean productivity by optimizing the discrete-valued dilution rate function. Based on duality theory, the established min–max discrete optimal control problem is first transformed into a single level minimization problem, which is then discretized into a large-scale parameter optimization problem with semi-infinite constraint via time transformation and control parameterization. A new two-step estimation of distribution algorithm is developed to solve the obtained large-scale optimization problem. Numerical results show the feasibility and effectiveness of the proposed solution approach together with the superiority of the obtained control strategy considering parameter uncertainties.

A flexible framework to coordinate debris clearance and relief distribution operations: A robust machine learning approach

This study aims to offer an integrated, flexible action plan to bridge two vital operations in the response phase of disaster management, namely debris clearance and relief items distribution. On this subject, a broad range of associated decisions, including location-allocation, inventory, scheduling, and routing, are reflected in a multi-period environment. Also, so as to provide a pliable decision chain, two substantial kinds of flexibility are considered, including network design and due dates. So, a new bi-objective mathematical model is proposed, where the first objective minimizes the total cost, and the second one minimizes the total operation time. Since travel time information accuracy has an undeniable influence on system performance, a new two-phase data-driven methodology, including a hybrid machine learning model and the distributionally robust optimization with φ- divergence, is offered to estimate reliable travel time and tackle its uncertainty. Additionally, the application and validation of the recommended model and solution methodology are investigated in a real case study. What is more, in order to explain the superiority of the proposed solution approach concerning robustness solution, various simulation experiments are executed. Ultimately, several numerical experiments are implemented, and captured results manifest that the proposed framework can enhance the system’s performance. The obtained results reveal that considering the flexibility in network design and delivery date reduces the total costs and operation time by 7 and 6 percent.

Learning efficient in-store picking strategies to reduce customer encounters in omnichannel retail

Omnichannel retailers are reinventing stores to meet the growing demand of the online channel. Several retailers now use stores as supporting distribution centers to offer quicker Buy-Online-Pickup-In-Store (BOPS) and Ship-From-Store (SFS) services. They resort to in-store picking to serve online orders using existing assets. However, in-store picking operations require picker carts traveling through store aisles, competing for store space, and possibly harming the offline customer experience. To learn picking policies that acknowledge interactions between pickers and offline customers, we formalize a new problem called Dynamic In-store Picker Routing Problem (diPRP). This problem considers a picker that tries to pick online orders (seeking) while minimizing customer encounters (hiding) – preserving the offline customer experience. We model the problem as a Markov Decision Process (MDP) and solve it using a hybrid solution approach comprising mathematical programming and reinforcement learning components. Computational experiments on synthetic instances suggest that the algorithm converges to efficient policies. We apply our solution approach in the context of a large European retailer to assess the proposed policies regarding the number of orders picked and customers encountered. The learned policies are also tested in six different retail settings, demonstrating the flexibility of the proposed approach. Our work suggests that retailers should be able to scale the in-store picking of online orders without jeopardizing the experience of offline customers. The policies learned using the proposed solution approach reduced the number of customer encounters by up to 50%, compared to policies solely focused on picking orders. Thus, to pursue omnichannel strategies that adequately trade-off operational efficiency and customer experience, retailers cannot rely on actual simplistic picking strategies, such as choosing the shortest possible route.

Scheduling of log logistics using a metaheuristic approach

An efficient log logistics plan results in log procurement cost savings. In practice, log logistics presents complexities such as synchronization of different machines, sorting of logs, and compatibility requirements. To address these complexities, this research proposes a decomposition approach for optimization of log logistics considering synchronization of log loaders and heterogeneous trucks at both cut blocks and sort yards. In the first phase, the daily number of truckloads between cut blocks and sort yards for each trucking contractor is determined, while allocation of the truckloads to compatible trucks and detailed routing and scheduling decisions are addressed in the second phase. Furthermore, a simulated annealing algorithm is developed for the second phase to obtain the schedules for large-sized problems. Additionally, the parameters of the algorithm are tuned using the Taguchi method. The algorithm is then applied to a case of a large Canadian forest company in British Columbia, Canada with a 1-month planning horizon. The results show that the proposed solution approach can successfully satisfy synchronization requirements and generate detailed schedules in a reasonable time. Also, cost savings for contractors are possible by assigning overtime rather than utilizing more trucks to fulfill their transportation tasks.

Issues and Challenges of Quality Assurance in Higher Education Institutes: A Systematic Literature Review

Quality Assurance is a timely topic in higher education that ensures the quality has been fulfilled in order to achieve the expected objectives. While many success stories and improvements have been reported, implementation issues and challenges have been highlighted in quality assurance systems. Therefore, a systematic literature review can be used to be aware of recent advancements and unsolved problems of the scenario. To this end, this study reviews the related research studies highlighting the issues, challenges, and proposed solution approaches. Accordingly, this study identified issues of quality assurance under three main categories: 1. institute-related issues, 2. quality assurance process related issues, and 3. quality assurance stakeholder related issues. Further, three main solution approaches were identified: 1. attitude changes and increasing the awareness of stakeholders, 2. policy-level changes of the system, and 3. quality assurance process improvements and facilitating information system solutions to address these issues. Among these solutions, process improvements and information system solutions can be identified as a substantial solution to address these issues. However, to create an effective information system solution to facilitate higher education quality assurance, a reliable model is necessary. This literature review highlights the lack of such a reliable model for developing an information system that aims to facilitate higher education quality assurance.

Blood supply chain configuration and optimization under the COVID-19 using benders decomposition based heuristic algorithm

During COVID-19, blood demand exceeded pre-pandemic levels due to reduced donations, causing shortages. Given the severe shortage, it's crucial to optimise blood use, prevent shortages, minimise wastage, and reduce unnecessary transfusions in all hospitalised patients. Designing a reliable blood supply chain network (BSCN) is an effective solution, especially for COVID-19 patients. This strategic decision significantly impacts emergency management performance. An efficient and reliable blood supply chain requires the consideration of multiple factors, including scarceness and perishability of blood, simultaneously. However, existing studies have not addressed all relevant factors in an integrated blood supply chain, and this paper aims to bridge this gap. Furthermore, an efficient Benders Decomposition based heuristic approach is proposed to solve the model. The solution approach has been compared with a set of commonly used meta-heuristic algorithms, including the red deer algorithm (RDA), tree growth algorithm (TGA), and genetic algorithm (GA). The outcomes illustrate that the proposed heuristic approach can solve small and large-size problems in significantly less CPU time than the other proposed solution approaches. For large-size problems, it can reduce the average CPU time by about 80% compared to TGA, about 80% compared to GA, and about 83% compared to RDA. A real case study has been implemented to validate the proposed mathematical model and solution method. The sensitivity analysis has been conducted to validate the significance of the model's parameters; consequently, several managerial insights have been derived.

Metro track geometry car routing problem with periodic demands: a case study in Beijing, China

Because of their curvature, age, track type and maintenance history, metro tracks should be inspected periodically during the planning horizon. Metro companies operate track geometry cars to periodically collect measurement data from the tracks by crossing the urban rail transit network. Optimizing the schedule for the track geometry car is a challenge: in addition to minimizing travelling distance, the inspection time interval of the same line should be arranged as equally as possible within the planning horizon. This article proposes a mathematical model for the track geometry car routing problem with periodic demands (TGCRP-PD). To effectively address this problem, a memetic algorithm-based metaheuristic is adopted. The proposed solution approach is applied to a real-world case. The numerical results show that the proposed method could save a dead mileage of 295.016 km (accounting for 48.882%) and greatly improve the level of service intervals.

Tramp ship routing and speed optimization with tidal berth time windows

Tramp shipping companies need to schedule shipping routes and decide on appropriate speeds based on short-term demand. This task differs from traditional vehicle routing problems (VRPs) in that the ship may wait for the tide, which changes with time. The wait time is a nonlinear function of the load, and in this paper, we describe this kind of wait as the ship following a tidal berth time windows. Additionally, the speed of the ship affects both the wait time and the sailing cost. This paper proposes a mixed-integer nonlinear programming model to tackle this problem. A branch-and-price framework is applied to solve the model efficiently, decomposing the model into a set partitioning master problem and an elementary shortest path subproblem. A labeling algorithm incorporating ship speed is developed to handle the subproblem, and further enhancements are made by optimizing the speed separately. Computational experiments show the effectiveness and accuracy of the proposed solution approach for large-scale instances. Moreover, considering the tidal time windows allows for exploiting the practical benefits of raising tides, which benefits the tramp shipping industry.

Demand Management for Peak-to-Average Ratio Minimization via Intraday Block Pricing

Price based demand response schemes may significantly improve power system efficiency. Additionally, it is desired that such schemes yield improved power operation, by reducing the peak consumption. This paper proposes the Intraday Block Pricing (IBP) scheme, aiming to promote effective demand response among consumers by charging their electricity usage based on intraday time-slots. To design the prices associated with the proposed scheme, we formulate a bilevel optimization problem that aims to minimize the Peak-to-Average Ratio (PAR) and simultaneously benefit the consumers and the utility company. The bilevel problem is converted into a single-level Mathematical Program with Equilibrium Constraints (MPEC). The resulting MPEC is non-convex and includes nonlinear constraints. Hence, to obtain a solution, it is relaxed into a Mixed Integer Linear Program by dealing with all nonlinearities. To evaluate the conservativeness of the proposed approach, a lower bound to the cost of the original bilevel problem is obtained. The applicability of the proposed scheme is demonstrated with simulations on various case studies, which exhibit a significant reduction in PAR and economic gains for the utility company and consumers. Moreover, simulation results show that the solutions of the original and relaxed problems are equivalent, demonstrating the effectiveness of the proposed solution approach. Further simulation results demonstrate significant advantages in the performance of the IBP scheme when compared to existing schemes in the literature.

A logic-based Benders decomposition algorithm for a repair crew routing and drone scheduling problem after a natural disaster

Natural disasters may lead to the isolation of rural areas due to the destruction of roads. Therefore, as a direct consequence of the impossibility of distribution of relief goods, more damages are expected in the aftermath of disasters in these areas. In this situation, we assume that damaged roads must be repaired immediately by a single repair crew and based on a specific route to prevent further damages and facilitate access to ground relief. There exists usually some stored medical equipment in isolated areas so that it can be consumed until destroyed roads become accessible again. In order to reflect this issue in this research, deadlines are considered for making isolated areas accessible. These deadlines are flexible and can be updated whenever a drone visits an isolated area and provide the area with limited amounts of air relief items. An integer linear programming is developed to formulate the repair crew routing problem and drone scheduling in an integrative manner. The developed model is complex and fails to provide the optimal solution even for instances including more than six demand areas and six damaged roads within one hour when the deadline is tight. Therefore, we developed a logic-based Benders decomposition algorithm that is able to find the optimal solution of instances including sixteen nodes in less than three quarters. Using a set of randomly generated instances, we compare the performance of the proposed solution approaches and provide a set of sensitivity analyses based on significant parameters of the problem.

A hybrid simulation and optimisation approach for capacity allocation of operating rooms under uncertainty

Operating rooms are the most important assets of hospitals whose capacity allocation to different specialties is critical and challenging. This paper proposes a hybrid simulation and optimisation approach for the strategic surgery capacity allocation problem over a yearly horizon. First, a simulation model is proposed to specify an appropriate percentage of the total available time of operating rooms in each period which is allocated to emergency patients. Then, a new mathematical model is developed to determine the optimal capacity plan (i.e. resource allocation) for operating rooms by handling both elective and emergency patients under uncertainty of surgery demands, surgery times and length of stays. The model aims to minimise the total under-time and the number of patients on the waiting list according to their surgery priorities. A tailored fix and relax algorithm is devised to solve the problem efficiently. Finally, a real case study in a private hospital along with some test problems are provided to evaluate the proposed model and its solution method. The numerical results demonstrate the desirable performance of the proposed solution approach. Moreover, the results indicate a significant reduction in the total under-time and the number of beds allocated in each period.

Variable neighborhood search-based algorithms for the parallel machine capacitated lot-sizing and scheduling problem

This paper addresses the capacitated lot-sizing and scheduling problem on parallel machines with eligibility constraints, sequence-dependent setup times, and costs. The objective is to find a production plan that minimizes production, setup, and inventory holding costs while meeting the demands of products for each period without delay for a given planning horizon. Since the studied problem is NP-hard, we proposed metaheuristic approaches, Variable Neighborhood Search, Variable Neighborhood Descent, and Reduced Variable Neighborhood Search algorithms to analyze their performance on the problem. Initially, we presented an initial solution generation method to satisfy each period’s demand. Then, we defined insert, swap, and fractional insert moves for generating neighborhood solutions. We employed an adaptive constraint handling technique to enlarge the search space by accepting infeasible solutions during the search. Lastly, we performed computational experiments over the benchmark instances. The computational results show the effectiveness of the proposed solution approaches, compared to existing solution techniques in the literature, and the improvements in various problem instances compared to the best-known results.

Solving crop planning and rotation problems in a sustainable agriculture perspective

In this paper, the problem of planning the allocation of crops to arable lands, taking into account crop rotations principles and diversification strategies promoted by sustainable agriculture is addressed. Optimization models for solving multi-period planning problems are proposed, able to decide how to allocate crops in each growing season in order to maximize the total expected profit. The allocation decisions are made considering the crop rotation benefits across seasons and the sustainable requirements stated by current regulations. A complexity analysis is performed, and polynomial special cases are presented. Integer Linear Programming models are proposed, for a case study related to structured professional Italian farms specialized in arable crops, following sustainability rules coming from public regulations and private initiatives, i.e., the Common Agricultural Policy by the European Union and “La Carta del Mulino” by the Barilla Group. Numerical experiments conducted on real data show the effectiveness of the proposed solution approaches.

Transient response of energy harvesting systems with multi-well potential under Poisson white noise excitations

This paper studies the transient response of vibration-based energy harvesting (VEH) systems characterized by multi-well potential functions exposed to Poisson white noise excitations. With the recently developed radial basis function neural networks (RBFNN) approach, the generalized Fokker–Planck–Kolmogorov (GFPK) equation in three-dimensional state space for the transient probability density function (PDF) is solved. The trial solution is composed of a series of radial basis Gaussian functions, each multiplied by a time-varying weight coefficient. The method of Lagrange multiplier is introduced for the constraint of weight coefficients to normalized the trial solution of the PDF solution. As examples to validate the proposed solution approach, the mono-stable and tri-stable energy harvester systems are studied. Monte Carlo simulations (MCS) have been applied to check the analytical solution. Some theoretical trends of the harvested mean square voltage (MSV) as a function of system parameters are examined. The non-Gaussian excitation and time-varying nature of the system can significantly affect the MSV output. It is noted that this work provides researchers and engineers a tool to optimize energy harvesting systems operated in random environment.

An efficient multi-agent approach to order picking and robot scheduling in a robotic mobile fulfillment system

This paper presents a Multi-Agent System (MAS) for optimizing the scheduling and routing of mobile robots and human pickers in a Robotic Mobile Fulfillment System (RMFS). The RMFS is a system designed for e-commerce warehousing where autonomous mobile robots are used to fetch inventory pods, also referred to as racks, from the storage area and transport them to the appropriate picking station where human pickers pick the required number of goods. The system requires the solution of several hard decision problems like: the order-to-picking station assignment and sequencing, the pod selection, and the multi-robot task allocation. The proposed solution approach employs decentralized scheduling mechanisms, i.e. auctions and dispatching rules, to communicate and distribute picking and retrieval tasks among the agents. Various dispatching rules are identified and analyzed over a wide set of problem instances of the RMFS with varying numbers of mobile robots, picking stations, and order sizes. The proposed MAS framework shows promising results and requires only a fraction of the computation time compared to a centralized scheduling algorithm. The MAS also includes both unidirectional and bidirectional lanes. Although additional complexity in collision avoidance is introduced when using bidirectional lanes, it allows for better system performance.