ࡱ > b V [ R bjbj I ΐ ΐ ) m " " k k T 3 3 3 3 U # =( =( =( =( 7* Z * * T T T T T T T $ lX [ 6 T * * " 7* * * T k k =( =( LU B B B * k R =( =( T B * T B B n M t " O =( E! 3 ; " YO mT bU 0 U yO h D[ > ~ D[ @ O D[ O * * B * * * * * T T )@ * * * U * * * * D[ * * * * * * * * * " 1 : ezS -NVR{|ST P 2 7 4 . 5 e.shƋxA WN9eۏD P S O {lvv^LKmՋNROS^xvz s!`1 e)Yg2 Ngewm1 1 . wmQ*zz]zf[byxq\NpS 2 6 4 0 0 1 2 . wmQ*zz]zf[bxvzu{t'Yq\NpS 2 6 4 0 0 1 Xdv^LKmՋNQ\KmՋeTMNOKmՋb,gv:_'YOR]b:NS_MRꁨRKmՋ|~SU\veT0[v^LꁨRKmՋǏz-NKmՋNR^ YBgNOSvNP S O {l:NW@xǏ[zzxv͑e[INv^Џ(uNS0S_{P[~QNev|P[MOnvfelQ_cQN Ny9eۏTvD P S O {l0Oncv^LKmՋ[begP[t~QNv^LKmՋNR^vvhQpeN~_gagN0Ng5uP[ňYv^LKmՋ|~-N NWW5ugv^LKmՋ:NO[9eۏvD P S O {lۏLNNw_0RN gO^KmՋ^R0~ghfNW O{lvk9eۏTvD P S O {lN!kpef\[O'`f}Y(uN]z^(u0 sQ.͋v^LKmՋNR^ gO^R9eۏvyce|P[OS{l R e s e a r c h o n o p t i m i z e T a s k S c h e d u l i n g f o r P a r a l l e l T e s t B a s e d o n I m p r o v e d D P S O A l g o r i t h m W a n g Y i p i n g 1 , W e n T i a n z h u 2 , L i W e n h a i 1 1 . N a v a l A e r o n a u t i c a l E n g i n e e r i n g I n s t i t u t e , D e p a r t m e n t o f S c i e n t i f i c R e s e a r c h , Y a n t a i 2 6 4 0 0 1 C h i n a 2 . N a v a l A e r o n a u t i c a l E n g i n e e r i n g I n s t i t u t e , G r a d u a t e S t u d e n t s B r i g a d e , , Y a n t a i 2 6 4 0 0 1 C h i n a A b s t r a c t : T h e p a r a l l e l t e s t h a s b e c o m e t h e d e v e l o p m e n t a l t r e n d o f A u t o m a t i c T e s t S y s t e m w i t h g r e a t s t r e n g t h i n r e d u c i n g t e s t t i m e a n d t e s t c o s t . A i m e d a t t h e p r o b l e m s t h a t t a s k s c h e d u l i n g i s c o m p l e x a n d t a s k o p t i m i z a t i o n is difficulty in parallel automatic test, a improved Discrete Particle Swarm Optimization (DPSO) algorithm is proposed, in which problem space coding is redefined and particle position update formula is rebuilt using crossover and mutation operator. And then the objective function and constraint condition of task scheduling for parallel test are given, according to the limit completion time theorem of parallel test. In order to validate the performance of the improved DPSO algorithm, a parallel test simulation experiment for three pieces of circuit board is made by parallel test system of certain radar electronic equipment, and the optimal task scheduling is got. The results show that compared with genetic algorithm the improved DPSO algorithm has less iterations, higher efficiency and better optimal performance, and is more suitable for engineering application. Key words: p a r a l l e l t e s t t a s k s c h e d u l i n g o p t i m i z e s e q u e n c e i m p r o v e d d i s c r e t e p a r t i c l e s w a r m o p t i m i z a t i o n ( D P S O ) a l g o r i t h m 0 _ v^LKmՋb/g/fS_MRꁨRKmՋ|~SU\eTKN N[ 1 ] 0[(WT NeQS[bYyNRvKmՋ[ 2 ] NcؚN|~vKmՋHesQ\NKmՋYvnev^Ǐ[DnvqQNeg~KmՋb,g[ 3 , 4 ] 0 (Wv^LKmՋ|~-NY*NKmՋNRv^S0W3uDnv^LЏLSONuDnzNNQz`QYUOubnKmՋؚHesBlvv^LKmՋ^R Nv/f N*N YBgv0NOSvN P [ 5 , 6 ] 0 1 P S O {li |P[OSP a r t i c l e S w a r m O p t i m i z a t i o n P S O {l ge/f(W1 9 9 5 t^1u>yO_tf[[K e n n e d y TV5ul]z^E b e r h a r t (We.s[ 7 ] -N@bcQv[/fۏS{lv N*NeR/e0 P S O {lHQ:gRYS|P[k*N|P[Nh N*NBlvO v^1uvhQpenx[ N*N^