Search

Internal audit is an independent function that provides objective assurance and advisory services to business partners in the form of projects. As the skilled human resources, internal auditors play a significant role in ensuring audit quality and thus should be managed properly. However, there are only very few studies on auditor-project assignment problems at the operational level. Also, many studies on general team formation problems neglect the real-life implications of simplifying mathematical models to enable practitioners to understand and implement. To bridge these gaps, this study presents a new integrated two-phase approach to assign internal auditors to multiple projects. In the designed decision support model, a harmonious level index is firstly proposed, measuring the suitability between auditor capability and project characteristics using a simplified fuzzy filter ranking method. Then, Min–Max multi-choice goal programming (MMCGP) model provides the assignment solution that balances internal auditors' multi-dimensional fit to the projects and their preferences, and thus optimizes the productivity of internal audit projects and staff's job satisfaction. A real case study is presented to illustrate the effectiveness of the proposed model and its solution. The comparison results further show that the proposed model outperforms the manual operations and the exiting approaches in the literature. This research extends the existing literature on auditor assignment problem by developing a flexible decision support tool and a practical classification of auditor competency. The MMCGP approach also provides an ideal staff assignment solution by setting multiple aspiration levels for each goal and balancing the unwanted deviations among goals. [ABSTRACT FROM AUTHOR]

In this research, we focus on the multi-mode resource-constrained multi-project scheduling problem from the perspective of consultants who attend design-bid-build contracts and put their efforts to optimize the client and contractors' profits simultaneously. Each contractor may have a different level of knowledge and experience to perform a project. Therefore, the contractor selection to execute a project may affect the quality and satisfaction level of the project. This study aims to schedule activities of a certain number of projects with multiple execution modes and pre-specified relations. A multi-objective mixed-integer linear programming model is proposed to maximize satisfaction in contractor selection, minimize the total cost of projects and earliness and tardiness penalties, and minimize the total completion time of projects subject to resource constraints. It is assumed that activity durations are uncertain; thereby the globalized robust counterpart of the formulation is presented such that the normal range of the perturbation is the intersection of a box and a polyhedral. The proposed multi-objective model is solved by the multi-choice goal programming approach. This problem is computationally complex and the Benders decomposition (BD) method is utilized to solve it. Numerical experiments on different instances prove the quality of the proposed BD method and illustrate that the proposed model is capable of selecting a proper contractor and multi-project scheduling. [ABSTRACT FROM AUTHOR]

Orthogonal Nonnegative Matrix Factorization (ONMF) with orthogonality constraints on a matrix has been found to provide better clustering results over existing clustering problems. Because of the orthogonality constraint, this optimization problem is difficult to solve. Many of the existing constraint-preserving methods deal directly with the constraints using different techniques such as matrix decomposition or computing exponential matrices. Here, we propose an alternative formulation of the ONMF problem which converts the orthogonality constraints into non-convex constraints. To handle the non-convex constraints, a penalty function is applied. The penalized problem is a smooth nonlinear programming problem with quadratic (convex) constraints that can be solved by a proper optimization method. We first make use of an optimization method with two gradient projection steps and then apply a post-processing technique to construct a partition of the clustering problem. Comparative performance analysis of our proposed approach with other available clustering methods on randomly generated test problems and hard synthetic data-sets shows the outperformance of our approach, in terms of the obtained misclassification error rate and the Rand index. [ABSTRACT FROM AUTHOR]

The integration of drones into the conventional truck delivery system has gained substantial attention in the business and academic communities. Most existing works restrict the launch and recovery of unmanned aerial vehicles (UAVs) to customer locations (or nodes) in the delivery network. Nevertheless, emerging technological advances can allow drones to autonomously launch/land from a moving vehicle. In addition, majority of the current literature assumes multiple UAVs to be deployed and/or recovered simultaneously, thereby ignoring the associated scheduling decisions, which are essential to ensure safe, collision-free operations. This research introduces the single truck multi-drone routing and scheduling problem with en route operations for last-mile parcel delivery. A mixed integer linear programming (MILP) model is developed to minimize the delivery completion time. In addition, a variant is introduced to minimize the total delivery cost. Since the problem under consideration is NP-hard, a relax-and-fix with re-couple-refine-and-optimize (RF-RRO) heuristic approach is proposed, where the associated decisions (truck routing and drone scheduling) are decomposed into two stages and solved sequentially. Besides, a deep learning-based clustering procedure is developed to establish an initial solution and accelerate the convergence speed of the RF-RRO heuristic. Notably, the proposed approach is extended to solve a multi-truck multi-drone variant using a deep learning-based cluster-first route-second heuristic. Our numerical results show that the proposed MILP model is able to solve problem instances with up to 20 customers optimally in a reasonable time. The proposed RF-RRO heuristic can achieve optimal (or near-optimal) solutions for small instances and is computationally efficient for large cases. Extensive experimental analysis shows 30% average savings in delivery completion time, and an average drone utilization of 62% if en route drone operations are considered. In addition, numerical results provide insights on the impact of heterogeneous drone fleet and customer density. [ABSTRACT FROM AUTHOR]

Faced with dynamic and increasingly diversified public transport requirements, bus operators are urged to propose operational innovations to sustain their competitiveness. In particular, ordinary bus operations are heavily constrained by well-established route options, and it is challenging to accommodate dynamic passenger flows effectively and with a good level of resource utilization performance. Inspired by the philosophy of sharing economy, many of the available transport resources on the road, such as minibuses and private vehicles, can offer opportunities for improvement if they can be effectively incorporated and exploited. In this regard, this paper proposes a metric learning-based prediction algorithm which can effectively capture the demand pattern and designs a route planning optimizer to help bus operators effectively deploy fixed routing and cooperative buses with traffic dynamics. Through extensive numerical studies, the performance of our proposed metric learning-based Generative Adversarial Network (GAN) prediction model outperforms existing ways. The effectiveness and robustness of the prediction-supported routing planner are well demonstrated for a real-time case. Further, managerial insights with regard to travel time, bus fleet size, and customer service levels are revealed by various sensitivity analysis. [ABSTRACT FROM AUTHOR]

Maintenance planning aims to improve the reliability of assets, prevent the occurrence of asset failures, and reduce maintenance costs associated with downtime of assets and maintenance resources (such as spare parts and workforce). Thus, effective maintenance planning is instrumental in ensuring high asset availability with the minimum cost. Nevertheless, to find such optimal planning is a nontrivial task due to the (i) complex and usually nonlinear inter-relationship between different planning decisions (e.g., inventory level and workforce capacity), and (ii) stochastic nature of the system (e.g., random failures of parts installed in assets). To alleviate these challenges, we study a joint maintenance planning problem by considering several decisions simultaneously, including workforce planning, workforce training, and spare parts inventory management. We develop a hybrid solution algorithm (DRLSA ) that is a combination of Double Deep Q-Network based Deep Reinforcement Learning (DRL) and Simulated Annealing (SA) algorithms. In each episode of the proposed algorithm, the best solution found by DRL is delivered to SA to be used as an initial solution, and the best solution of SA is delivered to DRL to be used as the initial state. Different from the traditional SA algorithms where neighborhood structures are selected only randomly, the DRL part of DRLSA learns to choose the best neighborhood structure to use based on experience gained from previous episodes. We compare the performance of the proposed solution algorithm with several well-known meta-heuristic algorithms, including Simulated Annealing, Genetic Algorithm (GA), and Variable Neighborhood Search (VNS). Further, we also develop a Machine Learning (ML) algorithm (i.e., K-Median) as another benchmark in which different properties of spare parts (e.g., failure rates, holding costs, and repair rates) are used as clustering features for the ML algorithm. Our study reveals that the DRLSA finds the optimal solutions for relatively small-size instances, and it has the potential to outperform traditional meta-heuristic and ML algorithms. [ABSTRACT FROM AUTHOR]

This study investigates an integrated airline fleet introduction and assignment problem under a daily route-based network. It is critical for an airline to integrate advanced digital technology with the methodology of operational research to achieve smart operations. An integer programming model is proposed for this problem. Based on the model, a hybrid algorithm combining a multi-encoding variable neighborhood search (VNS) and Branch-and-Bound (B&B) method is proposed. Firstly, the multi-encoding VNS solves fleet introduction and determines fleet numbers. A self-adaptive multi-layer encoding process with nine neighborhood structures is designed in the VNS process to improve the local search efficiency. Secondly, given the fleet numbers, the improved B&B method driven by profit matrix is designed to solve fleet assignment. There are three modified strategies in the B&B method containing cyclic best-first search, wide branching, and upper bound pruning. Finally, computational experiments are conducted to test algorithm performance. The proposed algorithm is compared with basic VNS, differential evolution, and genetic algorithm on objective value, running speed, and stability. The experiments show that the proposed model and algorithm offer a smart methodology for airline integrated problem. [ABSTRACT FROM AUTHOR]

We consider the two-machine permutation flow shop scheduling problem with uncertain job processing time, which is sampled from no specific distribution type. For the ease of discussion, an ambiguity set with a priori mean and support set information is constructed. We then introduce a distributionally robust optimization (DRO) perspective to handle the uncertainty. To the best of our knowledge, this is the first time that a DRO method is applied to this problem setting. Given that the original DRO model is nonlinear and intractable in nature, we first reformulate the inner maximization problem into a linear programming model with a fixed sequence, based on the duality theory and optimality conditions. By including the sequence decision, we further transform it into an equivalent mixed-integer linear programming (MILP) problem via incorporating the valid lower and upper bounds and McCormick inequalities. The obtained MILP could be solved to optimality with the off-the-shelf commercial solvers. In the numerical study, it is demonstrated that the DRO-based model could effectively solve the large scale instances with up to 100 jobs optimally within 30 s. Compared with the SLP, DRO model always triumphs on the worst-case indicator. And as the problem scale increases, the DRO model gradually outperforms the SLP in terms of the Up-90% and Up-75% indicators. Furthermore, the optimal sequence obtained by the deterministic model is less stable than the DRO model, which can enhance the robustness of the manufacturing system against processing uncertainty. [ABSTRACT FROM AUTHOR]

The main resources in software projects are human resources equipped with various skills, which makes software development a typical intelligence-intensive process. Therefore, effective human resource scheduling is indispensable for the success of software projects. The aim of software project scheduling is to assign the right employee to the right activity at the right time. Uncertainty is inevitable in software development, which further complicates the scheduling of projects. We investigate the software project scheduling problem with uncertain activity durations (SPSP-UAD) and aim at obtaining effective scheduling policies for the problem. We present and transform a scenario-based non-linear chance-constrained stochastic programming model into an equivalent linear programming model. To solve the NP-hard SPSP-UAD efficiently, we develop a hybrid meta-heuristic TLBO-GA that combines the teaching–learning-based optimization algorithm (TLBO) and the genetic algorithm (GA). Our TLBO-GA is also equipped with some problem-specific operators, such as population initialization, rows exchange and local search. We use simulation to evaluate the scheduling policies obtained by our algorithms. Extensive computational experiments are conducted to evaluate the performance of our TLBO-GA in comparison to the exact solver CPLEX and four existing meta-heuristic algorithms. The comparative results reveal the effectiveness and efficiency of our TLBO-GA. Our TLBO-GA provides an extensible and adaptive automated scheduling decision support tool for the software project manager in the complex and uncertain software development environment. [ABSTRACT FROM AUTHOR]

We consider distributed approval voting schemes. Each voter i ∈ I has α i cards that (s)he distributes among the candidates a ∈ A as a measure of approval. One (or several) candidate(s) who received the maximum number of cards is (are) elected. We provide polynomial algorithms to recognize voting forms and voting correspondences generated by such voting schemes in cases when either the number of candidates or the number of voters is equal to 2. We prove that for two voters, if α 2 ≥ α 1 - 2 ≥ 0 then the unique voting correspondence has distinct rows. We also characterize voting forms with distinct rows. [ABSTRACT FROM AUTHOR]

The uncertain flow is a common factor that leads to risks in hub-and-spoke systems. Confronted with risks, it cannot be ignored that the decision maker has different risk-bearing capacities, which influence decision-making. This paper presents an α -risk appetite to characterize the risk-bearing capacity and employs the concept of belief degree to construct it. A multi-commodity capacitated p-hub median model is built with time windows, uncertain flows and the α -risk appetite objective function. In this model, the uncertain flows are depicted as uncertain variables, which possess an empirical uncertainty distribution even with little data. Uncertainty theory has the advantage in converting the proposed uncertain model into a deterministic equivalent form. While the deterministic model is still a mixed 0–1 integer programming problem and has challenges in theory and practice for addressing large-scale problems. Based on the difficulty, we design a hybrid genetic algorithm, where a new principle of cis-position assignment is proposed and inserted into this algorithm to accelerate the solving process. By comparing the proposed model with a deterministic benchmark model, the computational results demonstrate that taking the α -risk appetite and uncertainty into account could avoid the losses of neglecting risks. The observations of cost and hub location reflect the model's sensitivity about risk appetite levels. Finally, the effectiveness of the proposed algorithm is verified by two data sets, the Civil Aeronautics Board data and the Australian Post data. [ABSTRACT FROM AUTHOR]

Operations in areas of importance to society are frequently modeled as mixed-integer linear programming (MILP) problems. While MILP problems suffer from combinatorial complexity, Lagrangian Relaxation has been a beacon of hope to resolve the associated difficulties through decomposition. Due to the non-smooth nature of Lagrangian dual functions, the coordination aspect of the method has posed serious challenges. This paper presents several significant historical milestones (beginning with Polyak's pioneering work in 1967) toward improving Lagrangian Relaxation coordination through improved optimization of non-smooth functionals. Finally, this paper presents the most recent developments in Lagrangian Relaxation for fast resolution of MILP problems. The paper also briefly discusses the opportunities that Lagrangian Relaxation can provide at this point in time. [ABSTRACT FROM AUTHOR]

The class social network is a momentous factor when it comes to educational, personal and professional student success as well as achieving course learning outcomes. Students and teachers benefit from expanded network connectivity via augmented engagement, more inclusivity, and efficient diffusion of information. We present a novel method for positively influencing the class social network. We develop an in-class grouping strategy based on optimization and sociocentric network analysis that pragmatically expands the students' social networks. In contrast to existing routines, our technique focuses on maximizing individual student opportunities to establish new ties. Based on the knowledge of existing connections, our procedure systematically optimizes the overall number of new ties that can be established during a team project. Our data-driven approach is designed for practical use in class. We show that the underlying combinatorial problem of maximizing unrelated intra-team students can be modeled as a bin packing variant. Using an integer programming formulation, we demonstrate the efficient spreadsheet implementation. We discuss model extensions to account for high-density networks, team balancing, and teammate forcing and forbidding, allowing for hybridization using existing grouping techniques. In an empirical study, we provide evidence for the efficacy of our approach using data from 10 industrial engineering classes with 253 students and 77 project teams - in both face-to-face and virtual modes. We demonstrate the impact of our grouping method compared to random-assignment, self-selection, and maximizing existing intra-team ties. We report an impressive 62% increase of ties compared to only 17% when self-assigning. [ABSTRACT FROM AUTHOR]

As the only largest source of revenue and cost in a hospital, the operation room (OR) scheduling problem is a hot research topic. Nonetheless, an integrated model is the missing key to managing and improving the efficiency of ORs. This paper presents a fully integrated model regarding three concepts: meditating elective patients and emergency patients together, considering ORs and downstream units, and proposing hierarchical weekly, daily, and rescheduling models. Due to the inherent randomness in emergency patient arrival, a random forest machine learning model and geographical information systems are used to obtain the emergency patient surgery duration and arrival time, respectively. According to the machine learning model in weekly and daily scheduling, initially, fixed capacity is reserved for emergency patients. When an emergency patient arrives, the surgery starts if a reserved OR is available. Otherwise, the first available OR will be dedicated to the patient due to an emergency patient's higher priority than an elective patient. In this case, it is needed to reschedule the OT schedule for the remaining patient. Moreover, the three-phase model guarantees that an emergency patient assigns to an OR within a specific time limit. To solve the models, genetic algorithm and particle swarm optimization are developed and compared. In addition, a real-world case study is undertaken at a hospital. The results of comparing the proposed approach to the hospital's current scheduling show that the three-phase model had a considerable positive effect on the ORs schedule. [ABSTRACT FROM AUTHOR]

The multi-objective integer programming problem often occurs in multi-criteria decision-making situations, where the decision variables are integers. In the present paper, we have discussed an algorithm for finding all efficient solutions of a multi-objective integer quadratic programming problem. The proposed algorithm is based on the aspect that efficient solutions of a multi-objective integer quadratic programming problem can be obtained by enumerating ranked solutions of an integer quadratic programming problem. For determining ranked solutions of an integer quadratic programming problem, we have constructed a related integer linear programming problem and from ranked solutions of this integer linear programming problem, ranked solutions of the original integer quadratic programming problem are generated. Theoretically, we have shown that the developed method generates the set of all efficient solutions in a finite number of steps, and numerically we have elaborated the working of our algorithm and compared our results with existing algorithms. Further, we have analyzed that the developed method is efficient for solving a multi-objective integer quadratic programming problem with a large number of constraints, variables and objectives. [ABSTRACT FROM AUTHOR]