Jovan B. Popovic

Decision making on stock levels in cases of uncertain demand rate

Abstract Inventory control is a typical problem of decision making. In this paper a periodic replenishment of stock, the spare parts being of one kind, is discussed for some cases when the demand rate is uncertain. The first decision, before all others in the sequence, is done by assuming an a priori distribution of demand rate. In time, as the demand process goes on, corrections of parameters of the a priori distribution are made according to the accumulated knowledge about past demand. This Bayesian approach to decision making based on learning about the uncertain demand rate is known for the case when the demand rate is unknown but constant. It is shown that this same approach can be used in some cases when the demand rate is unknown and not constant. Results are given and used for inventory control.

Dynamic programming—neural network real-time traffic adaptive signal control algorithm

In this paper, an “intelligent” isolated intersection control system was developed. The developed “intelligent” system makes “real time” decisions as to whether to extend (and how much) current green time. The model developed is based on the combination of the dynamic programming and neural networks. Many tests show that the outcome (the extension of the green time) of the proposed neural network is nearly equal to the best solution. Practically negligible CPU times were achieved, and were thus absolutely acceptable for the “real time” application of the developed algorithm.

Vehicle routing in the case of uncertain demand: a Bayesian approach

The “classical” stochastic vehicle routing problem includes the stochasticity of the demands but presupposes that the distributions of the demands at the nodes are known. In this paper we assume that the distributions of the demands at the nodes are not known. This assumption brings even more uncertainty to a problem which is already complex. In practice, when such a problem is faced, then, almost as a rule, the real problem is transformed into a “convenient” one by neglecting the fact that the distributions are not known, i.e. by introducing some appropriate distributions which make the problem solvable. This paper suggests how to solve the real problem taking into account the uncertainty, i.e. the fact that the distributions at the nodes are not known. To deal with the uncertainty the Bayesian approach is used.

An adaptive method for generating demand inputs to airline seat inventory control models

In order to increase the passenger load factor and total profit, most airlines offer several different fares to passengers on the same flight. Seat inventory control is the process of assigning the available number of seats to different fare classes. During the last three decades, various aspects of this problem have been considered in the literature. The basic input data for the existing airline seat inventory models are the demand distributions for seats in particular fare classes. Statistical data as a base for forecasting future passenger demand are very often not available (opening a new route, significant changes in fares, the appearance of competitive carriers, etc.). The current models are not able to handle cases where no prior historical data is available. This paper attempts to develop an approach to deal with this problem. The adaptive airline seat inventory control, which is proposed, is based on the Bayesian approach. The model developed is illustrated by numerical examples.

Chronology of the development of geodetic reference networks in Serbia

In this paper, the development of geodetic reference networks in Serbia is shown. This historical summary covers the time from the first organised work in 1855 until today. Special attention has been paid to the establishment of the modern network using Global Navigation Satellite Systems (GNSS). Within the networks established by GNSS, two last realised networks are especially distinguished. The first one is a classic spatial reference network SREF (Serbian Reference Frame) established in 2003, and the second one is a permanent stations network – AGROS (Active Geodetic Reference Network of Serbia) realised in 2006.