Simulation-based Optimization Approach Research Articles

A continuous review policy for two complementary products with interrelated demand

We consider a retailer selling two complementary products. The products are demanded individually or jointly. The retailer uses a continuous review (Q,r) policy to manage the inventory of each product. The customers arrive according to a Poisson process and unmet demands are lost. We aim to calculate the optimal policy parameters which maximize the expected profit rate. A shortage of one product might result in a lost sale of its complementary product, even if this product is in stock. This fact complicates the analysis. We rely on a simulation-based optimization approach to derive the optimal parameters of the (Q,r) policy for each product. Our results reveal that the maximum profit rate is attained when all customers demand both products, i.e., the bundle of complementary products. In addition, we observe that when the demand rate is low, the interaction between product demands has significant impact on the expected profit rate, and this effect becomes even more prominent when the unit lost sale costs are low and the unit holding costs are high. We derive the analytical properties of a single-product inventory system based on which an efficient approximation algorithm is developed.

Simulation-based optimization for modeling and mitigating tunnel-induced damages

This research develops a simulation-based optimization approach that is capable of modeling and mitigating tunnel-induced damages. Two fuzzy cognitive maps (FCMs) (i.e., one with self-feedback and the other without self-feedback) are learned from historical datasets by using the real-coded genetic algorithm (RCGA) on a data-driven modeling manner. Then, the optimal variable value set is searched in the input space. Two new measures, namely “maximum response” and “average response”, are proposed to search the optimal variable value set in the input space in the FCM dynamic simulation process. A realistic tunnel case in the Wuhan metro system in China is extensively investigated to demonstrate the applicability and effectiveness of the developed approach. Results indicate that (1) The FCM with self-feedback is more stable than the FCM without self-feedback considering its higher coefficient of determination in the testing samples, where less modification of input variables realizes comparable improvement in the objective in the FCM with self-feedback. (2) The measure “maximum response” shows a larger change in the objective than the measure “average response”, where modifications in input space are similar. (3) It is revealed that the ground settlement is more sensitive to TBM operational parameters than tunnel geometry and geological conditions in the two learned FCMs. The developed approach provides insights into a better understanding of causal relationships among factors in tunnel-induced damages, enabling the planning of proactive control strategies for mitigating tunnel-induced damages.

Aggregation–Decomposition-Based Multi-Agent Reinforcement Learning for Multi-Reservoir Operations Optimization

Stochastic dynamic programming (SDP) is a widely-used method for reservoir operations optimization under uncertainty but suffers from the dual curses of dimensionality and modeling. Reinforcement learning (RL), a simulation-based stochastic optimization approach, can nullify the curse of modeling that arises from the need for calculating a very large transition probability matrix. RL mitigates the curse of the dimensionality problem, but cannot solve it completely as it remains computationally intensive in complex multi-reservoir systems. This paper presents a multi-agent RL approach combined with an aggregation/decomposition (AD-RL) method for reducing the curse of dimensionality in multi-reservoir operation optimization problems. In this model, each reservoir is individually managed by a specific operator (agent) while co-operating with other agents systematically on finding a near-optimal operating policy for the whole system. Each agent makes a decision (release) based on its current state and the feedback it receives from the states of all upstream and downstream reservoirs. The method, along with an efficient artificial neural network-based robust procedure for the task of tuning Q-learning parameters, has been applied to a real-world five-reservoir problem, i.e., the Parambikulam–Aliyar Project (PAP) in India. We demonstrate that the proposed AD-RL approach helps to derive operating policies that are better than or comparable with the policies obtained by other stochastic optimization methods with less computational burden.

Managing EMS systems with user abandonment in emerging economies

In many emerging economies, callers may abandon ambulance requests due to a combination of operational (small fleet size), infrastructural (long travel times) and behavioral factors (low trust in the ambulance system). As a result, ambulance capacity, which is already scarce, is wasted in serving calls that are likely to be abandoned later. In this article, we investigate the design of an ambulance system in the presence of abandonment behavior, using a two-step approach. First, because the callers’ actual willingness to wait for ambulances is censored, we adopt a Maximum Likelihood Estimator estimation approach suitable for interval censored data. Second, we employ a simulation-based optimization approach to explicitly incorporate customers’ willingness to wait in: (i) tactical short-term decisions such as modification of dispatch policies and ambulance allocations at existing base locations; and (ii) strategic long-term network design decisions of increasing fleet size and re-designing base locations. We calibrate our models using data from a major metropolitan city in India where historically 81.3% of calls were successfully served without being abandoned. We find that modifying dispatch policies or reallocating ambulances provide relatively small gains in successfully served calls (around 1%). By contrast, increasing fleet size and network re-design can more significantly increase the fraction of successfully served calls with the latter being particularly more effective. Redesigning bases with the current fleet size is equivalent to increasing the fleet size by 8.6% at current base locations. Similarly, adding 29% more ambulances and redesigning the base locations is equivalent to doubling the fleet size at the current base locations and adding 34% more ambulances and redesigning base locations is equivalent to a three-fold increase. Our results indicate that in the absence of changes in behavioral factors, significant investment is required to modify operational factors by increasing fleet size, and to modify infrastructural factors by redesigning base locations.

Online Parameter Estimation Methods for Adaptive Cruise Control Systems

Two new methods are presented for estimating car-following model parameters using data collected from the Adaptive Cruise Control (ACC) enabled vehicles. The vehicle is assumed to follow a constant time headway relative velocity model in which the parameters are unknown and to be determined. The first technique is a batch method that uses a least-squares approach to estimate the parameters from time series data of the vehicle speed, space gap, and relative velocity of a lead vehicle. The second method is an online approach that uses a particle filter to simultaneously estimate both the state of the system and the model parameters. Numerical experiments demonstrate the accuracy and computational performance of the methods relative to a commonly used simulation-based optimization approach. The methods are also assessed on empirical data collected from a 2019 model year ACC vehicle driven in a highway environment. Speed, space gap, and relative velocity data are recorded directly from the factory-installed radar unit via the vehicle's CAN bus. All three methods return similar mean absolute error values in speed and spacing compared to the recorded data. The least-squares method has the fastest run-time performance, and is up to 3 orders of magnitude faster than other methods. The particle filter is faster than real-time, and therefore is suitable in streaming applications in which the datasets can grow arbitrarily large.

Effect of load bundling on supply Chain inventory management: An evaluation with simulation-based optimisation

ABSTRACT In this paper, the effect of load bundling on overall costs, service level, and CO2 emissions is evaluated for a multi-stage, multi-item supply chain. A simulation-based optimisation approach is used to optimize the inventory management parameters reorder point and lot size. The optimisation approach consists of a simulation model and a metaheuristic search procedure, which is a subclass of the evolutionary algorithm. For the evaluation of the load bundling opportunity in different demand structures, a multicriterial objective function is optimized. The paper shows that the load bundling opportunity has significant cost and environmental benefits. The study points out that the load bundling opportunity leads to smaller and more customer-driven lot sizes which simultaneously reduce the carbon emissions. Finally, results show that for medium to high service level target values, ABC-clustered order rate scenarios lead to lower supply chain costs than demand scenarios with an identical order rate for all items.

A risk-averse simulation-based approach for a joint optimization of workforce capacity, spare part stocks and scheduling priorities in maintenance planning

We model a maintenance system consisting of one repair facility, where repairables are kept on inventory to serve assets to prevent downtime and increase availability. We seek optimal values of the repairable spare parts stocks and workforce capacity in the repair facility. Further, we simultaneously search for the best repair scheduling rule that minimizes total inventory holding and backorder costs associated with the downtime of assets. The joint optimization problem under study brings about two additional challenges: (i) the difficulty of analyzing such systems due to the lack of analytical (i.e., queuing) models, and (ii) the difficulty in incorporating the decision maker’s risk attitude regarding uncertainties. We develop a risk-averse simulation-based optimization approach, in which the decision maker’s risk attitude is modeled as a trade-off between the expected and the worst-case costs in the objective function. In the developed approach, the repairable spare part supply system is analyzed with a discrete-event simulation (DES) model. The DES model is coupled with an improved reduced variable neighborhood search (IRVNS) meta-heuristic that seeks the optimal values of decision variables. We compare the performance of the proposed risk-averse simulation-based optimization approach with several plausible benchmark methods commonly used in practice and with well-known meta-heuristic algorithms.

Formulation and solution approach for calibrating activity-based travel demand model-system via microsimulation

This study addresses the problem of calibrating utility-maximizing nested logit activity-based travel demand model-systems. After estimation, it is common practice to use aggregate measurements to calibrate the estimated model-system’s parameters prior to their application in transportation planning, policy making, and operations. However, calibration of activity-based model-systems has received much less attention. Existing calibration approaches are myopic heuristics in the sense that they do not consider the fundamental inter-dependencies among choice-models and do not have a systematic way to adjust model parameters. Also, other purely simulation-based approaches do not perform well in large-scale applications. In this study, we focus on utility-maximizing nested logit activity-based model-systems and calibrating aggregate statistics such as activity shares, mode shares, time-dependent & mode-specific OD flows, and time-dependent & mode-specific sensor counts. We formulate the calibration problem as a simulation-based optimization problem and propose a stochastic gradient-based solution procedure to solve it.The solution procedure relies on microsimulation to calculate expectations of the aggregate statistics of interest to the calibration problem. Additionally, we derive approximate analytical expressions for the gradient of the objective function —that are evaluated through microsimulation on mini-batches of the population. The proposed solution procedure is sensitive to the fundamental structure of the activity-based model-system and is non-myopic in considering the dependencies across its model components. The formulated optimization problem is non-convex, highly nonlinear, and potentially has multiple-minima. Finally, we show —through a real-world application— that the proposed solution procedure outperforms other state-of-the-art purely simulation-based optimization approaches in terms of computational efficiency, stability, and convergence. We also compare various gradient-based solution algorithms to determine the best algorithm to update the parameters. This work has the potential to facilitate wider and easier application of activity-based model-systems.

Cascade utilization of LNG cold energy by integrating cryogenic energy storage, organic Rankine cycle and direct cooling

Utilizing LNG cold energy in different temperature ranges with distinctive approaches is a promising option to achieve a high thermodynamic efficiency. This paper proposed a novel LNG cold energy cascade utilization (CES-ORC-DC-LNG) system by integrating cryogenic energy storage (CES), organic Rankine cycle (ORC), and direct cooling (DC) to recover LNG cold energy in the low, middle, and high temperature ranges, respectively. The optimal components and their respective fractions of the mixed working fluids in ORCs, and other key parameters were simultaneously determined by a simulation-based optimization approach which was developed in this study. Then, comprehensive energetic and exergetic analyses were conducted to compare the proposed system performance with the system in the literature to highlight the superior thermodynamic performance. The results indicated that the net power output of the CES-ORC-DC-LNG system was 103.30 kW, which was 22.48% higher than the base case. The optimal compositions of the mixed working fluids in ORC-1 (ethane: 23.73%, ethylene: 15.01%, R134a: 20.17%, and R152a: 41.09%) and ORC-2 (ethylene: 60.14%, R134a: 38.73%, R1270: 0.54%, and R134a: 0.59%) were distinctive with the net power output of 13.89 and 14.46 kW, respectively. The round trip efficiency of the proposed system was 141.88% with 19% increment compared with the base case. Furthermore, the exergetic efficiency of the CES-ORC-DC-LNG system was 73.92% which was 5.17% higher than the base case. Thus, the CES-ORC-DC-LNG system is an energy efficiency LNG cold energy cascade utilization system and can be a potential technology for large-scale cryogenic energy storage.

Joint optimization of production, quality control and maintenance for serial-parallel multistage production systems

This paper presents a joint model of production, quality control and preventive maintenance for a serial-parallel multistage production system. In each stage, there are multiple machines to meet the productivity and line-balance requirement. The machines deteriorate with usage and thus influences the product quality. A make-to-stock production policy is adopted to provide protection to stock against uncertainties. All items processed in each stage undergo quality checks. Conforming items are delivered to the next stage for further processing, whereas non-conforming items are scrapped. During production runs, based on the quality information feedback, preventive maintenances are performed to improve the machines reliability and hence the product quality. At the end of production runs, machines are inspected and maintained if necessary. The novelty of the proposed maintenance strategy consists in that both the machine structure importance measure and productivity are considered when selecting machines for maintenance. It aims to simultaneously determine the length of production run, quality control-threshold and maintenance-threshold such that the average cost rate is minimized. A stochastic mathematical model is developed which is solved by a simulation-based optimization approach coupling Monte Carlo Simulation and genetic algorithm. Finally, an illustrative example and some comparisons are provided to demonstrate the model.

Mitigating Cascading Outages in Severe Weather Using Simulation-Based Optimization

Severe weather events can trigger cascading power outages and lead to significant losses. In this work, we investigate cascading outage mitigation under severe weather conditions. Given day-ahead weather forecasts and component failure models, we aim to identify a set of power lines that can be hardened to minimize the expected impact of potential cascading outages. Since the expected load shedding cannot be expressed as an explicit function of line hardening decisions and system states, we developed a cascading outage simulator to estimate the expected value of load shedding under various initial weather-related disruption scenarios generated using a weather forecast. To avoid massive enumeration of all possible combinations of line hardening decisions and reduce the simulation efforts, we employed an efficient simulation-based optimization approach that quickly identifies the (near) optimal line hardening decisions in the presence of both large simulation noises due to the highly variable initial disturbances and system states, and significant randomness in the subsequent cascades. The algorithm is also able to utilize parallel computing to dramatically reduce computation time to support decision making in preparation for severe weather conditions. We performed a case study on the Northeast Power Coordinating Council (NPCC) 140-bus system model to demonstrate that our approach can significantly improve power grid resilience to adverse weather events.

Facility location and capacity planning considering policy preference and uncertain demand under the One Belt One Road initiative

Abstract Due to its importance in integrating facilities and smoothing supply chain, facility location and capacity planning problem (FLCPP) is no doubt a key issue in constructing an efficient logistics and transportation network. One Belt and One Road (OBOR) initiative aims at reconstructing the logistics and transportation network along the trade corridors and improving the trade connectivity among the developing and developed countries. Nevertheless, under the OBOR initiative, FLCPP is greatly influenced by the government policy for regional development and the uncertain demand incurred by the enormous difference of economic form, custom and culture awareness among the involved countries. This study focuses on a special FLCPP under the OBOR initiative, in which the uncertainties of both policy preference and customer demand are considered simultaneously. To cope with these two categories of uncertainties, this study proposes a simulation-based optimization approach that includes the rough simulation and fine simulation methods. The former aims at finding a set of feasible decisions by using fuzzy and stochastic simulation, while the latter is used to achieve the best decision from the obtained feasible decisions by hybridizing the stochastic simulation with optimal computing budget allocation and mathematical programming. In order to demonstrate the application and advantage of the proposed approach, simulation experiments are conducted and experimental results show that this approach performs well in dealing with the concerned problem as well as some managerial insights and policy suggestions are obtained according to the analysis and results.

Capacity planning in the hospital queueing network with blocking: simulation-based optimization approach

Capacity planning in the hospital queueing network with blocking: simulation-based optimization approach

Designing a stochastic multi-objective simulation-based optimization model for sales and operations planning in built-to-order environment with uncertain distant outsourcing

To this day, sales & operations planning (S&OP) remains one of the most critical challenges of supply chain management in terms of improving customers’ experience level and managing supply chain costs. Nowadays the synchronization of sales essentials and industrial restrictions in an uncertain environment is critical for all global companies facing complexity in supply chain networks and operations systems. Sales and supply chain structures in build-to-order industries generally have cross-functional and conflicting objects that are not easily comparable. The current research is an extension of an innovative S&OP model within a more dynamically complex system in terms of demands uncertainty and unreliability of backup supplies. The promoted model is utilized for more efficient management and integration of industrial restrictions and sales essentials in global supply chain networks in terms of demand and supply coordination and responsiveness toward customers’ needs. Accordingly, this research has applied a simulation-based optimization approach to balance the trade-off between logistics cost and customers’ experience level as the main objectives. Two classes of strategies are offered for handling parts’ replenishment and flexibility to the sales structure. Initially, the proposed model evaluates two main inventory strategies’ performances. Thereafter, the optimized values of flexibility degree and safety stock fraction for corresponding inventory strategies are calculated by using a multi-objective simulation-optimization method. The investigation begins with an overall assessment of the system's performance. Furthermore, it details the alterations among the strategies’ key performance indicators (KPIs) as system parameters’ function.

Design of Wake Equalizing Ducts using RANSE-based SBDO

We propose a Simulation-Based Design Optimization (SBDO) approach for the design of an Energy Saving Device (ESD) based on the Wake Equalizing Duct (WED) concept. Pre-Ducts, like Wake Equalizing Ducts, reduce the wake losses, improve the propeller-hull interaction and generate an additional thrust. An integrated design approach, relying on a parametric description of the duct geometry of the WEDs and on a RANSE method, managed by a global convergence optimization algorithm, is developed to maximize the delivered thrust. The Japan Bulk Carrier (JBC) test case, for which model scale experimental data on the effectiveness of the WED are available in the literature, is considered as a baseline. Results obtained by using the adaptive, fully automated proposed design framework highlight significant improvements of the overall propulsive efficiency when the Pre-Duct design is tailored to the actual hull wake shape. In addition, off-design conditions are considered to verify the robustness of the proposed designs with respect to variations of the working point and discuss the opportunity of a robust optimization process.

Using simulation-based system dynamics and genetic algorithms to reduce the cash flow bullwhip in the supply chain

The bullwhip effect (BWE) is a phenomenon, which is caused by ineffective inventory decisions made by supply chain members. In addition to known inefficiencies caused by the bullwhip effect within a supply chain product flow, such as excessive inventory, it can also lead to inefficiencies in cash flow such as the cash flow bullwhip (CFB). The CFB reduces the efficiency of the supply chain (SC) through heterogeneous distribution of cash among supply chain members. This paper aims to decrease both the BWE and the CFB across a SC through applying a simulation-based optimisation approach, which integrates system dynamics (SD) simulation and genetic algorithms. For this purpose, cash flow modelling is incorporated into the SD structure of the beer distribution game (BG) to develop the CFB function. A multi objective optimisation model is then integrated with the SD-BG simulation model. Finally, a genetic algorithm (GA) is applied to determine the optimal values for the inventory, supply line, and financial decision parameters. Results show that the proposed integrated framework leads to efficient liquidity management in the SC in addition to cost management.

A Simulation-Based Optimization Approach for Reliability-Aware Service Composition in Edge Computing

With the prevalence of Internet of Things (IoT), edge computing has emerged as a novel computing model for optimizing traditional cloud computing systems by moving part of the computational tasks to the edge of the network for better performance and security. With the technique of services computing, edge computing systems can accommodate the application requirements with more agility and flexibility. In large-scale edge computing systems, service composition as one of the most important problems in services computing suffers from several new challenges, i.e., complex layered architecture, failures and recoveries always in the lifecycle, and search space explosion. In this paper, we make an attempt at addressing these challenges by designing a simulation-based optimization approach for reliability-aware service composition. Composite stochastic Petri net models are proposed for formulating the dynamics of multi-layered edge computing systems, and their corresponding quantitative analysis is conducted. To solve the state explosion problem in large-scale systems or complex service processes, time scale decomposition technique is applied to improving the efficiency of model solving. Additionally, simulation schemes are designed for performance evaluation and optimization, and ordinal optimization technique is introduced to significantly reduce the size of the search space. Finally, we conduct experiments based on real-life data, and the empirical results validate the efficacy of the approach.

Simulation-Based Optimization Approach to Multi-Choice Transportation Problem

Simulation-Based Optimization Approach to Multi-Choice Transportation Problem

A long-term fleet renewal problem under uncertainty: A simulation-based optimization approach

In this paper, we model and solve a strategic problem of fleet renewal to meet future operational needs under uncertain conditions. The fleet renewal problem focuses on mainly strategic decisions involving from fleet size, fleet mix and timing of replacement, yet it is essential to consider a significant amount of detail regarding short-term decisions to prevent inferior or infeasible strategies. In this direction, we develop a hybrid simulation model by combining system dynamics (SD) and discrete event simulation (DES) approaches. The standalone use of this model enables the decision maker to analyze the effects of both short- and long-term decisions on availability by simulating the processes that the fleet undertakes through its life-cycle from asset acquisition to retirement. Nevertheless, the simulation neither suggests nor seeks the best renewal strategy(ies). To alleviate this difficulty, we propose a simulation-based optimization that uses a genetic algorithm (GA) to effectively search a very large set of feasible fleet renewal strategies and uses the developed hybrid simulation model to evaluate candidate strategies found by GA. To provide a decision context where the approach has been developed and applied, we use a naval fleet renewal application. The extensive numerical experiments show that the proposed approach not only finds good and robust renewal strategies but also identify critical resources that influence the fleet’s availability. Finally, the robustness of optimized strategies under uncertainty is tested by sensitivity analysis, and mappings between implemented strategies and the fleet performance are constructed by scenario discovery analysis to provide insights for decision makers.

Production control of failure-prone manufacturing-remanufacturing systems using mixed dedicated and shared facilities

This paper addresses a control problem within hybrid manufacturing-remanufacturing systems (HMRS) evolving in a dynamic and stochastic environment. Unlike previous works, it contributes to the discourse by considering mixed dedicated and shared facilities (MDSF). It is motivated by the need to integrate the remanufacturing flow while increasing the capacity of the system to react to uncertainties (random failures and repairs of facilities, rates of returned products, etc.). The problem is to determine the production rates for the manufacturing and remanufacturing facilities as well as the setup decisions required to switch the remanufacturing facility from a mode supplied by returned products to a mode supplied by raw materials. The objective is to minimize the expected total cost. A dynamic stochastic model is thus proposed, and the developed optimality conditions are solved numerically. The obtained control policy combines hedging point policies and a stock-based setup strategy, under which switching is based on the finished products stock, according to threshold rules. Such control policy is then analyzed and compared to the literature by considering three different control policies. Two of those were developed in other contexts using only dedicated facilities or shared facilities in a dynamic and stochastic environment while the other is adapted to ours. For an effective comparison study, a simulation-based optimization approach is adopted and implemented for the problem under study. Extensive numerical results suggest that the use of MDSF managed by the proposed control policy gives the best performance in terms of costs.