Forecasting Air Astana’s spare parts and their inventory management

  • Tsakalerou, Maria (PI)
  • Papadopoulos, Chrysoleon (PI)
  • Babai , Mohamed-Zied (Other Faculty/Researcher)
  • Tsadiras , Athanasios (Other Faculty/Researcher)
  • O'Kelly, M. E J (Other Faculty/Researcher)
  • Turgunbayev, Yernar (Other Faculty/Researcher)

Project: Monitored by Research Administration

Project Details

Grant Program

Faculty Development Competitive Research Grant Program 2018-2020

Project Description

Purpose of the project is to conduct a thorough analysis of the spare parts demand faced by Air Astana, the National Airline company of the Republic of Kazakhstan, and to test a number of appropriate forecasting and inventory models in order to improve the current situation by predicting demand more accurately and implement inventory policies, respectively, which will have as an effect the reduction of the average total inventory costs of the company. Apart from the solution of this very challenging scientific problem, which is the primary goal of this research project, another objective of equal importance is the training of 24 Master second year students of the Master of Engineering Management (MEM) program (3 years X 2 groups X 4 students) within the context of their capstone projects on these two important problems of Supply Chain Management at a very large company. It is our belief that these students after their graduation will be highly qualified for a very good job as Materials Planners/Managers in any large and medium size company.
At a first phase, the Aircraft On Ground (AOG) spare parts are going to be considered and at a second phase, the remaining spare parts are also going to be examined. AOG spare parts are those which if they are missing cause an aircraft to stay on ground since it cannot fly till the respective spare parts are found and replaced. The cost of such an event is very high both for the airline and the passengers as the latter experience an unpleasant situation: delays in finding another aircraft, staying possibly at a hotel, among others (see attached pdf file by Draxler and Dzunda: conference paper, on the effect of AOG on the airlines).

Scientific novelty and significance.
A number of forecasting methods applicable to the sporadic (intermittent or irregular) demand are to be tested and validated with the data of Air Astana. These methods include: Croston’s method, the Boylan-Syntetos Approximation, the Bootstrapping method as well as Neural Networks, among others. The latter two methods are expected to give encouraging results, based on a preliminary research that has been already conducted within the context of the Capstone Projects of a group of last year MEM students and an extended research followed by the PI (Prof. Chrysoleon Papadopoulos, NU) and his partner/member of the research team (Associate Professor Athanasios Tsadiras, Aristotle University of Thessaloniki, Greece) with the assistance of the Materials Manager of Air Astana (Mr. Yernar Turgunbayev). Various modifications of the above and other methods (given in Section 3, below) are to be tested to see whether they fit better to the data of Air Astana. These may lead to new contributions in the area of forecasting and inventory models in the Aviation industry/Sector.
Definition of sporadic or intermittent or irregular demand: Intermittent demand or ID (also known as sporadic or irregular demand) comes about when a product experiences several periods of zero demand. Often in these situations, when demand occurs it is small, and sometimes highly variable in size. ID is often experienced in industries such as aviation, automotive, defence and manufacturing; it also typically occurs with products nearing the end of their life cycle. Some companies operating in these areas observe ID for over half the products in their inventories. In such situations there is a clear financial incentive to inventory control and retaining proper stock levels, and therefore to forecasting demand for these items.

The impact of the results on the development of science and technology and the expected social and economic impacts.
The impact of the results of this research is that Air Astana is going to see more accurate forecasting methods which will assist their Materials managers to apply inventory policies that will lead to less inventory costs for the company and simultaneously higher level of service to their passengers. As a Professor at NU, I count a lot on the educational and training aspect of this research project and the fact that a number of up to 24 Master students of the Master of Engineering Management (MEM) program of NU are to be educated and trained in the use and applicability of these forecasting methods and inventory models which will assist them in finding a job at Air Astana and other airlines or many other companies of Kazakhstan and abroad which face similar demand patterns (sporadic/intermittent/irregular) as well as any other company which carries inventory and needs to manage their stock in an efficient and effective way. Forecasting and Inventory management constitute a significant part of the Supply Chain Management and Operations Management in any company worldwide.

A review of previous research conducted in the world related to the topic under the study and their relationship with this project.

Daniel Waller from University of Lancaster (see attached pdf file, report) reviewed a number of forecasting methods for intermittent demand. These methods are described below in Section 3.
Ghobbar and Friend (2003, see attached pdf file) provided an evaluation of forecasting methods for intermittent parts demand in the field of aviation. A predictive model was proposed. Teunter and Duncan (2009, see attached pdf file) conducted a comparative study of forecasting methods for intermittent demand.
Effective start/end date3/20/188/30/21


  • Engineering management
  • Operations research
  • Forecasting and inventory models


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.