Demand forecasting is very often used in production planning, especially, when a manufacturer needs in a longer production cycle to respond flexibly to market demands. Production based on longer-term forecasts means bearing the risk of forecast unreliability in the form of finished product inventory deficit or excess. The use of computer simulation allows us to improve the planning process and optimise the plan for the intended goal. This paper presents the use of quantitative forecasting and computer simulations to create the production plan. Two approaches to production plan creation are demonstrated in a model case study. Products are characterized by varying demand and are produced on a single production line in continuous operation. The first approach uses ARIMA(2,0,2) (Auto-Regressive Integrated Moving Average) prognostic method selected as the most reliable method based on MAPE (Mean Absolute Percent Error). The second method applies Monte Carlo simulations and optimisation. The aim of the plan optimisation is minimisation the total costs connected with line rebuilding and storage of products. The comparison of the two approaches shows that planning using computer simulations and optimisation leads to lower total costs.
The paper presents quantitative approach for management decisions of the manufacturing system for production of fireplaces, related to evaluation of key parametersproductivity and throughput, which most authors and methodologies consider to be substantial. Methodology was based on creating the simulation model of the fireplace production line in software Witness; optimizing the production capacity by selecting constraints, based on results from simulation model; evaluating the simulation experiments with the goal to increase productivity; setting production to maximize sales profits using Simplex method. Simulation model was built according to a technological process of fireplaces in a semi-automated production. Improvement in a production process within theory of constraints philosophy is complemented by mathematical modelling -Simplex method, that estimate profit maximization in case the company management decides to produce more product variants.
This paper presents a methodology for the selection of an optimal investment variant using Monte Carlo simulation and OptQuest optimization. The decision-making process also includes risk analysis. Investment variants involve renewal and development of production equipment. Two approaches to investment decision making are introduced. The first approach is based on the analysis of the distribution function of Net Present Value (NPV), and the rule of mean value and coefficient of variation is used as the decision criterion for determining the profitability of investment variants. The second approach, based on the cumulative probability distribution of NPV, provides a comparative assessment of the investment variants using stochastic dominance rules. Both approaches lead to the choice of the same investment variant. In order to increase the profitability of the selected investment variant and reduce its risk, OptQuest optimization is subsequently implemented. The introduced approaches can be a useful support tool in investment decision-making.
AbstractTraffic modelling and simulation is one of the frequent tools used in road infrastructure design. Software tools designed for traffic simulations are an important supportive tool in decision-making and in choosing the optimal solution. The aim of this paper is to introduce the application of the VISSIM program, in the design and testing a model of the traffic-light-controlled intersection. The traffic on the selected congested intersection is modelled and simulated first for the current state, then for two models with modifications that are to increase the throughput of the intersection. The monitored criterion of the intersection throughput is the length of queues. Both adjustments have led to a significant reduction of the number of vehicles waiting in direction of the greatest congestions. In the first model, the average line length was reduced by 75%, and in the second model, the modifications lead to a fluent passage of right-turn vehicles and a significant reduction in vehicle lines for other directions.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.