English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

This article adopts partial least squares regression method to forecast the price of domestic (China-made) large aircraft C919. W ith analysis of the pricing background and the relationship between performance index and price of foreign aircraft, we build preliminary model. And comparing with other modeling methods by calculating average relative error, we find the preliminary model fits better. Considering the difference in environment between domestic and foreign market, we propose the concept of discount factor. A nd the factor is equal to 0.6529 by analyzing domestic and foreign small aircrafts of the sam e type. Finally through modified model we get the forecast price, namely 48.95 million dollars per aircraft. Th e forecasting model and discount factor have some reference value to the price forecast of domestic large aircraft. Introduction C919 is Chinese self-developed large aircraft. The project of it started in November, 2008. Its first flight is expected in 2014. And it's planned to obtain airworthiness certificate and put into service in 2016. When developing a new aircraft, we should consider lots of econom ic problems besides technique details. Among them, the questions whether and how long the airc raft manufacturer can recover the original investment outstand. To answer these questions, price forecast of the new aircraft is necessary. Partial least squares regression (hereinafter referred to as PLSR) can screen independent variables briefly. Its rich aided analysis technique can achieve screening while modeling and overcom e bad influence caused by variables' multiple correlation . In literature [2,3] by Li Shouan etc(2005,2006), the comparison among PLSR, regression without screening variables and stepwise multiple regression shows that PLSR is m ore accurate in price forecast of m ilitary aircraft than the other two models. Compared with price forecast relied on manufacturing cost and period expenses, the similar aircraft price forecast model based on analysis of military aircraft in literature [4] by Li Shouan et c (2007) is better. The price forecast background of the domestic large aircraft C919 is very similar to military aircraft. They both have a small number of sample data and a large number of price driving factors, as well as having few types. Besides, neither of their data collection is sound and their cost of development and manufacturing isn't available to the public. In literature [5] by LI Zhanke etc (2011), PLSR is adopted to forecast the price of C919, but the value of equalization-engineering-value-ratio is much too subjective and brief. In this article, we also apply PLSR to forecast the price of C919. And combining with China's actual situation, we deal with domestic small aircraft with PLS R, calculate discount factor, get modi fied model and obtain reasonable forecast price in ensured market competitiveness. Through the above work, we expect to provide some guidance for the price forecast of domestic large aircraft. 0022 2nd International Conference on Electronic & Mechanical Engineering and Information Technology (EMEIT-2012) Published by Atlantis Press, Paris, France. © the authors Introduction of Regression Model General linear relationship and log-linear relatio nship are selected to analyze here. To choose better model, we compare the fitting precision of the two methods. General linear regression form is as follows 0 1 1 2 2  n n y x x x α α α α = + + + + (1) Log-linear regression form is as follows 0 1 1 2 2 ln ln  n n y x x x β β β β = + + + + (2) Considering modeling problem with single dependent variable y and p independent variables 1 2 , ,, p x x x , the basic practice of PLSR is shown below. Primarily, extract the first component 1 t ( 1 t is the linear combination of 1 2 , ,, p x x x , and contains as much original information as possible. It’s required that th e relevance between 1 t and y reaches the m aximum. Then establish the regression equation of dependent variables y and 1 t . If its precision is satisfying, then the practice ends. Otherwise the extraction of the second component is needed, and it continues until reaching the required precision. Given that m components 1 2 , ,, m t t t are extracted ultimately, PLSR will establish the regression equation of y and 1 2 , ,, m t t t , and then tran sform it into th e regression equation of y and original independent variables, namely PLSR equation. The Preliminary Model of PLSR C919 is large aircraft, so regardless of performance or pricing, it's more reasonable to refer to other large commercial aircrafts' data than small aircrafts' and helicopters. A variety of factors considered, we choose the performance index and prices of 26 commercial aircrafts from the Boeing Company and Airbus as the basic data of PLSR modeling. The aircrafts' types and specific data are shown in the following table 1. Table 1 26 Large Commercial Aircrafts' Types and Specific Data Aircraft Type Capacity [Number of people] Span [m] length [m] Height [m] Maximum fuel load [kg] Maximum takeoff weight [kg] Maximum range [km] Maximum engine thrust [kN] Average price [million] 737-600 110 34.3 31.2 12.6 22000 65090 6000 91.6 55 737-700 126 34.31 33.6 12.55 22000 60330 6230 116 64 737-800 162 34.31 39.5 12.55 22000 70553 5665 124 76.75

The Price Forecast Model of China-made Large Aircraft Based on Partial Least Squares Regression