Jayendran Venkateswaran

Hybrid system dynamic—discrete event simulation-based architecture for hierarchical production planning

Multi-plant production planning problem deals with the determination of type and quantity of products to produce at the plants over multiple time periods. Hierarchical production planning provides a formal bridge between long-term plans and short-term schedules. A hybrid simulation-based hierarchical production planning architecture consisting of system dynamics (SD) components for the enterprise level planning and discrete event simulation (DES) components for the shop-level scheduling is presented. The architecture consists of the Optimizer, Performance Monitor and Simulator modules at each decision level. The Optimizers select the optimal set of control parameters based on the estimated behaviour of the system. The enterprise-level simulator (SD model) and shop-level simulator (DES model) interact with each other to evaluate the plan. Feedback control loops are employed at each level to monitor the performance and update the control parameters. Functional and process models of the proposed architecture are specified using IDEF. The internal mechanisms of the modules are also described. The modules are interfaced using High Level Architecture (HLA). Experimental results from a multi-product multi-facility manufacturing enterprise demonstrate the potential of the proposed approach.

Hierarchical production planning using a hybrid system dynamic-discrete event simulation architecture

Hierarchical production planning provides a formal bridge between long-term plans and short-term schedules. A hybrid simulation-based production planning architecture consisting of system dynamics (SD) components at the higher decision level and discrete event simulation (DES) components at the lower decision level is presented. The need for the two types of simulation has been justified. The architecture consists of four modules: enterprise-level decision maker, SD model of enterprise, shop-level decision maker and DES model of shop. The decision makers select the optimal set of control parameters based on the estimated behavior of the system. These control parameters are used by the SD and DES models to determine the best plan based on the actual behavior of the system. High level architecture has been employed to interface SD and DES simulation models. Experimental results from a single-product manufacturing enterprise demonstrate the validity and scope of the proposed approach.

Impact of modelling approximations in supply chain analysis – an experimental study

This paper presents a study of the comparison of the quality of results obtained at different levels of detail using a supply chain simulation. Analysis of supply chain is typically carried out using aggregated information to maintain the level of complexity of the simulation model at a manageable level. Advances in simulation have provided the ability to build comprehensive (detailed), modular models. The quantitative effect of detailed modelling on the corresponding analysis is investigated in this paper. A three-echelon supply chain is analysed using simulation models of varying levels of detail. Using each of these models, four sets of intensive experiments are performed. The first experiment intends to test whether the supply chain dynamics themselves depend on the modelling accuracy that represents the supply chain. The second and third experiments are conducted to test whether the effectiveness of the strategies employed to reduce the supply chain dynamics vary depending on the type (different detail) of model representing the supply chain. In the fourth experiment, statistical techniques are employed to identify which modelling aspect has the most influence on the supply chain dynamics. It is found that the approximations used in modelling, such as delays and capacity, have more impact on the outcome of supply chain analysis than end customer demand. Evidence that both the basic problem (supply chain dynamics) and the solution (strategy to reduce the dynamics) are greatly influenced by the modelling accuracy are presented.

Service levels, system cost and stability of production–inventory control systems

Production-inventory control system models have been analysed in the literature either in terms of their stability against demand fluctuations or in terms of their service level and cost performance under uncertain demand. This article analyses the production-inventory system performance in terms of service level (i.e. order fill rate) and average system costs, under stable settings of the control parameters. The classical automatic pipeline variable inventory and order-based production control system has been modified by explicitly modelling safety stock to help achieve higher services levels in the face of random demand. The stability of the system is affected by the control parameters: fractional rates of adjustment of work-in-process and inventory. However, the service level and average cost are affected by the control parameters as well as the smoothing factor in demand forecasting. This article puts forward five propositions which give light to general system performance based on the parameters selection. Intensive simulation experiments have also been carried out to reveal the performance variations within the stable region, leading to further insights on the system behaviour. The managerial insights which can assist proper tuning of systems to help achieve the desired performances have been discussed.

Distributed optimisation method for multi-resource constrained scheduling in coal supply chains

We consider an integrated planning and scheduling problem motivated by the coal supply chains in Australia. The problem considers production planning of several independent mines. The mines need trains to complete delivery of coal by the arrival of ships at the terminal. The trains, on the other hand, are in limited supply and therefore the mines need to share this common resource. For this problem, we present a mixed integer programming formulation which minimises total weighted earliness, tardiness and operational costs. We also present a distributed algorithm based on the Lagrangian relaxation, which incorporates the volume and Wedelin algorithms. The strength of our distributed algorithm is demonstrated by an extensive computational experiment on several randomly generated instances.

Hierarchical supply chain planning architecture for integrated analysis of stability and performance

In this paper, a novel architecture that allows a multi-scale federation of interwoven simulations and decision models to support integrated analysis of stability and performance in hierarchical production planning for supply chain networks is proposed. The proposed scheme is divided into four stages: plan stability analysis, plan optimisation, schedule optimisation and concurrent decision evaluation. The plans are more robust against future disturbances and more feasible for implementation than those from the conventional methods. Detailed functional and process models of the proposed system are specified using formal IDEF tools. The proposed approach is demonstrated using a realistic three-echelon supply chain system.

Distributed simulation: an enabling technology for the evaluation of virtual enterprises

This paper presents an application of distributed simulation to the evaluation of virtual enterprises. Each company or candidate can use a simulation of its facilities to determine if it has the capability to perform its individual function in the virtual enterprise. Then, these simulations can be integrated into a distributed simulation of the complete enterprise, and used to predict the viability and profitability of the proposed product collaboration. In this paper, a prototype distributed simulation for such a purpose is presented. First, information flows as well as material flows among members in a virtual enterprise are identified using IDEFO, a formal function modeling method. Sequences of the identified functions are then presented using the finite state automata formalism. These interactions are then implemented for a commercial simulation package. Finally, a distributed simulation composed of three individual simulations is successfully tested across platforms over both the internet and the local area network.

A hybrid simulation approach to planning in a VMI supply chain

In this paper, we propose an innovative approach of integrating the vendor managed inventory (VMI) strategy with a hierarchical approach to production planning (HPP) decisions within a supply chain environment. The proposed architecture is divided into three stages: plan optimisation (Stages I), schedule optimisation (Stage II) and decision evaluation (Stage III). To implement this architecture, we use system dynamics simulations, discrete-event simulations and the High Level Architecture (HLA) as a distributed infrastructure. We describe our approach for the analysis of supply chains in detail and discuss results from experiments that demonstrate the benefits of the proposed architecture.

Development and benchmarking of an epoch time synchronization method for distributed simulation

In this paper, a new epoch time synchronization approach for distributed simulation federates is presented. The approach allows federates in the simulation system to advance their local times at full speed while it is safe to do so. That is, the simulation moves rapidly to the minimum next epoch (interaction) event time, which is calculated using the minimum sojourn time for each federate, and then slows for federation synchronization. The proposed approach is demonstrated using a manufacturing supply chain simulation composed of four distributed federates. Experiments are executed to benchmark the proposed epoch time synchronization method against conventional conservative synchronization methods to show typical improvements for simulation operation. The experimental results reveal that the proposed approach reduces supply chain simulation execution time significantly while maintaining complete accuracy as compared with traditional conservative federation coordination approaches.

A resource constrained scheduling problem with multiple independent producers and a single linking constraint: A coal supply chain example

This paper examines a resource constrained production planning and scheduling problem motivated by the coal supply chain. In this problem, multiple independent producers are connected with a resource availability (or, linking) constraint. A general description of such problems is provided, before decomposing the problem into two levels. In the first level, we deal with production planning and in the second level, we deal with tactical resource scheduling. A real-world coal supply chain example is presented to anchor the approach. The overall problem can be formulated as an integrated mixed integer programming model which, in several cases, struggles to find even a feasible solution in reasonable amount of time. This paper discusses a distributed decision making approach based on column generation (CG). Computational experiments show that, the CG scheme has significant advantages over the integrated model and a Lagrangian relaxation scheme proposed by Thomas et al. (2013). This paper concludes with detailed discussions on the results and future research directions.