Research Project: Üretken4.0: Dijital Sanayinin Ve Ürün Yaşam Döngüsü Boyunca Optimize Edilmiş Tedarik Zinciri Yönetiminin Elektronik Ve Bilgi İletişim Teknolojileri İle Hayata Geçirilmesi
Loading...
Contributors
Funders
ID
TB.00329
Authors
Karaesmen, Fikri
Faculty Member
Publications
Supervised learning-based approximation method for single-server open queueing networks with correlated interarrival and service times
(Taylor _ Francis, 2021) Khayyati, Siamak; Tan, Barış; Department of Business Administration; Department of Industrial Engineering; Graduate School of Sciences and Engineering; Yes; College of Administrative Sciences and Economics; College of Engineering; GRADUATE SCHOOL OF SCIENCES AND ENGINEERING
Efficient performance evaluation methods are needed to design and control production systems. We propose a method to analyse single-server open queueing network models of manufacturing systems composed of delay, batching, merge and split blocks with correlated interarrival and service times. Our method (SLQNA) is based on using a supervised learning approach to determine the mean, the coefficient of variation, and the first-lag autocorrelation of the inter-departure time process as functions of the mean, coefficient of variation and first-lag autocorrelations of the interarrival and service times for each block, and then using the predicted inter-departure time process as the input to the next block in the network. The training data for the supervised learning algorithm is obtained by simulating the systems for a wide range of parameters. Gaussian Process Regression is used as a supervised learning algorithm. The algorithm is trained once for each block. SLQNA does not require generating additional training data for each unique network. The results are compared with simulation and also with the approximations that are based on Markov Arrival Process modelling, robust queueing, and G/G/1 approximations. Our results show that SLQNA is flexible, computationally efficient, and significantly more accurate and faster compared to the other methods.
Modelling and analysis of the impact of correlated inter-event data on production control using Markovian arrival processes
(Springer, 2019) Tan, Barış; Dizbin, Nima Manafzadeh; Department of Business Administration; Department of Industrial Engineering; Graduate School of Business; Yes; College of Administrative Sciences and Economics; College of Engineering; GRADUATE SCHOOL OF BUSINESS
Empirical studies show that the inter-event times of a production system are correlated. However, most of the analytical studies for the analysis and control of production systems ignore correlation. In this study, we show that real-time data collected from a manufacturing system can be used to build a Markovian arrival processes (MAP) model that captures correlation in inter-event times. The obtained MAP model can then be used to control production in an effective way. We first present a comprehensive review on MAP modeling and MAP fitting methods applicable to manufacturing systems. Then we present results on the effectiveness of these fitting methods and discuss how the collected inter-event data can be used to represent the flow dynamics of a production system accurately. In order to study the impact of capturing the flow dynamics accurately on the performance of a production control system, we analyze a manufacturing system that is controlled by using a base-stock policy. We study the impact of correlation in inter-event times on the optimal base-stock level of the system numerically by employing the structural properties of the MAP. We show that ignoring correlated arrival or service process can lead to overestimation of the optimal base-stock level for negatively correlated processes, and underestimation for the positively correlated processes. We conclude that MAPs can be used to develop data-driven models and control manufacturing systems more effectively by using shop-floor inter-event data.
An integrated data-driven method using deep learning for a newsvendor problem with unobservable features
(Elsevier, 2022) Karaesmen, Fikri; Pirayesh Neghab, D.; Khayyati, S.; Department of Industrial Engineering; Yes; College of Engineering
We consider a single-period inventory problem with random demand with both directly observable and unobservable features that impact the demand distribution. With the recent advances in data collection and analysis technologies, data-driven approaches to classical inventory management problems have gained traction. Specially, machine learning methods are increasingly being integrated into optimization problems. Although data-driven approaches have been developed for the newsvendor problem, they often consider learning from the available data and optimizing the system separate tasks to be performed in sequence. One of the setbacks of this approach is that in the learning phase, costly and cheap mistakes receive equal attention and, in the optimization phase, the optimizer is blind to the confidence of the learner in its estimates for different regions of the problem. To remedy this, we consider an integrated learning and optimization problem for optimizing a newsvendor's strategy facing a complex correlated demand with additional information about the unobservable state of the system. We give an algorithm based on integrating optimization, neural networks and hidden Markov models and use numerical experiments to show the efficiency of our method. In an empirical experiment, the method outperforms the best competitor benchmark by more than 27%, on average, in terms of the system cost. We give further analyses of the performance of the method using a set of numerical experiments.
Supervised-learning-based approximation method for multi-server queueing networks under different service disciplines with correlated interarrival and service times
(Taylor _ Francis, 2021) Khayyati, Siamak; Tan, Barış; Department of Business Administration; Graduate School of Sciences and Engineering; Yes; College of Administrative Sciences and Economics; GRADUATE SCHOOL OF SCIENCES AND ENGINEERING
Developing efficient performance evaluation methods is important to design and control complex production systems effectively. We present an approximation method (SLQNA) to predict the performance measures of queueing networks composed of multi-server stations operating under different service disciplines with correlated interarrival and service times with merge, split, and batching blocks separated with infinite capacity buffers. SLQNA yields the mean, coefficient of variation, and first-lag autocorrelation of the inter-departure times and the distribution of the time spent in the block, referred as the cycle time at each block. The method generates the training data by simulating different blocks for different parameters and uses Gaussian Process Regression to predict the inter-departure time and the cycle time distribution characteristics of each block in isolation. The predictions obtained for one block are fed into the next block in the network. The cycle time distributions of the blocks are used to approximate the distribution of the total time spent in the network (total cycle time). This approach eliminates the need to generate new data and train new models for each given network. We present SLQNA as a versatile, accurate, and efficient method to evaluate the cycle time distribution and other performance measures in queueing networks.
New product introductions with selection of unique and common features in monopoly markets
(Pergamon-Elsevier Science Ltd, 2024) Çelik, Burak; Tan, Barış; Schwarz, Justus Arne; Department of Business Administration; Yes; College of Administrative Sciences and Economics; College of Engineering
Firms have to determine the right features and prices for their new products as they introduce new product generations to the market. We consider the problem of determining the features of a new product that a monopolist will introduce into a market that contains an existing product as well as setting the prices of the existing and the new products. The firm also decides on offering only the new product or both the existing and the new products. We explicitly capture the effects of unique features, which are specific to one of the two products, and common features which are shared between the new and the existing product on these decisions. The problem is formulated as a nonlinear-mixed-integer program with general cost, demand, and price functions. For the case of linear cost, demand, and price, the nonlinear-mixed-integer program is converted to a nonlinear program and solved analytically. Based on this solution, the optimal prices for both products and the optimal unique features for the new product are derived in closed form, a linear-time algorithm is presented to determine the optimal common features, and the optimality conditions of keeping the existing product in the market are characterized. We show that the selection of the unique features, but not the common ones, is based on the difference between a feature's contribution to the product's demand and its cost adjusted by the price sensitivity in the linear case. Moreover, we find that the firm, if it wants to avoid demand cannibalization, should remove the existing product from the market rather than offer two products with mainly unique features. Capturing the effects of unique and common features directly allows firms to decide on the best rollover strategy and determine the right features and prices
