Dagstuhl Seminar 16062 Modeling and Analysis of Semiconductor Supply Chains Schloß Dagstuhl, 66687 Wadern, Germany Schloß Dagstuhl, 66687 Wadern, Germany 07-12, February 2016 Inventory Management in e to y a age e t Semiconductor Supply Chains Jei-Zheng Wu 1,3 , Chen-Fu Chien 2,3 , and Hui-Chun Yu 3,4 1 Department of Business Administration Soochow University Taiwan 1 Department of Business Administration, Soochow University, Taiwan 2 Department of Industrial Engineering and Engineering Management, National Tsing Hua University, Taiwan 3 NTHU-TSMC Center for Manufacturing Excellence, Taiwan 4 Taiwan Semiconductor Manufacturing Company, Taiwan 09 F b 09, February 2016 2016 1
Outline Basic Model not comprehensive Basic Model, not comprehensive Survival Analysis for Safety Stock Level (Inventory Days) Inventory Age and Accounting Principles Inventory Age and Accounting Principles 2
Semiconductor Manufacturing Process The rapid technology development and shortening p gy p g product life cycle lead to high risk of product obsolescence. Manufacturers still need to hold a reasonable level of inventory to satisfy customer under demand uncertainty and long lead-time. This study considers the work-in-process and inventory problem of the semiconductor memory manufacturing which has the 3- of the semiconductor memory manufacturing which has the 3 month approximately averaged lead time. General Specific p f (50 days) (5 days) (7 days) (22 days) 3
Features Complex BOM and product substitution, but also various raw wafer release schedules (wafer start), turnaround times (TAT, lead times), production plans, safety stock strategies, and end lead times), production plans, safety stock strategies, and end product-demand forecasts. Inventory on-hand and multi-period rolling mechanism Postponement strategy, delayed differentiation die-bank, wafer bank, via bank. • Although some manufacturers have set inventory banks Although some manufacturers have set inventory banks through the manufacturing process. However, the stock points do not bring its role into full play. It merely reflects the recent situation but lacks decisional meaning It l fl t th t it ti b t l k d i i l i • The sale unit will not provide the accurate distribution of demand. Inventory planners have to judge the demand. 4
Inventory Days Research Motivation and Objectives Research Motivation and Objectives Existing study did not suggest inventory days (turnover) for Existing study did not suggest inventory days (turnover) for decision making Inventory turnover is practical but only applies for post- y p y pp p evaluation in practice. This study aims to constructed an inventory-day-based multistage inventory decision model for calculating the inventory level of all of the stages of semiconductor manufacturing based on the i inventory survival analysis. t i l l i 5
Indices and Sets i, j product group i, j N the set of network structure of FG t Day t corresponding to DB m Month m NN NN the set of non network structure of FG the set of non-network structure of FG k technology k corresponding to DB l technology capacity balance level l DBN the set of DB existing in network T the set of days structure M M th the set of months t f th DBNN the set of DB existing in non-network K the set of technologies structure DM tm the set of day corresponding to month the set of product group f BOM FG FG the set of inventory group of stage FG the set of inventory group of stage FG corresponding to the group g of its corresponding to the group g of its DB the set of inventory group of stage DB upper stage in the bill of material WB the set of inventory group of stage WB the set of product group g MAP VB the set of inventory group of stage VB y g p g corresponding to its technology k corresponding to its technology k L the set of technology capacity balance level FL the set of FG corresponding to l 6
Parameters tat i turn-around time of product group i g i y i yield of product group i b it beginning on-hand inventory of product group i at day t it by it yielded beginning on-hand inventory of product group i at day t r i gross die of product group i i d it demand of product group i at day t ssld it inventory days of product group i at day t t ssl it safety stock level of product group i at day t c km capacity limit of technology k of month m bcp i inventory in the beginning of CP technology of product group i bd l technology capacity balance bound of level l rt l technology capacity balance ratio of level l 7
Variables end on-hand inventory of the period of product group i at day t end on hand inventory of the period of product group i at day t I it I it amount of shortage of product group i at day t S it receipt inventory from upper stage of product group i at day t F it unmet amount of safety stock of product group i at day t U it over amount of safety stock of product group i at day t O it estimated demand of product group i at day t i t D D it ti t d d d f d t t d the amount of work-in-process (WIP) of product group i at day t WIP it the amount of FG demand of product group i (corresponding to group j ) the amount of FG demand of product group i (corresponding to group j ) H ijt H ijt in network structure estimated safety stock level of product group i at day t SSL it technology capacity k used of month m C km inventory in the beginning of CP technology of product group i at day t BCP it technology capacity balance factor of product group i at balance level l i l l PL PL il t h l it b l f t f d t t b l l 8
Constraints relative to the Finished Goods Bank (FG) Finished Goods Bank (FG) Flow balance of safety stock level and the end on-hand inventory Fl b l f f t t k l l d th d h d i t , , ssl d t T i FG , , it is ssl O I U t T i FG it it it it i t s t ssld it it Estimated safety Stock flow from the last period into this period stock level , 1 1 , FG FG by b F F S S d d I I t t i i By summing the B i th it it it it it demand of product , 1 , by F S I d I S t T i FG group at DB stage ( 1 ) ( 1 ) it it it i t it it i t in the period of in the period of 0 , , 0 F i FG t tat inventory days up it i 9
Demand of the product group at Die Bank group at Die Bank - alternative BOM j i Network structure Network structure - substitutions • Multi-Chip Package - multi-chip , H H F F ijt ijt it ,( , ) , , t T i j BOM i FG j DB / , , D H y t T j DBN , jt ij t tat i i ( , ) i j N j i Non-network structure • Single-Chip Package g p g / , , D F y t T j DBNN , jt i t tat i i ( , ) i j NN 10
Constraints relative to the Die Bank (DB) the Die Bank (DB) Flow balance of safety stock level y and the end on-hand inventory T , , SSL O I U t i DB it it it it , , 1 b BCP F D I i DB t i it it it it , , , , 1 1 DB DB BCP BCP F F I I D D I I i i t t ( ( 1 1 ) ) it it it it i i t t it it it it 0 , , F i DB t tat it i Estimated safety stock level Estimated safety stock level Equation of beginning on-hand Equation of beginning on hand inventory of circuit probing process By summing the demand of at DB stage product group at DB stage in the period of inventory days up i d f i t d , , ( , ) y bcp BCP i DB i k MAP i i it T , , t tat SSL D t i DB k it is 0 0 , , BCP BCP i i DB DB t t t t tat i t s t ssld it i it 11
Constraints relative to the Wafer Bank (WB) the Wafer Bank (WB) Stock flow from the last Demand of the product group at WB period into this period / , , D F y t T j WB jt is i ( ( , ) ) i i j j BOM BOM s s t t tat tat , , 1 i WB by F D I i t it it it it Flow balance of safety stock level , , , , 1 by y F I D I i WB t ( ( 1 1 ) ) it it it it i i t t it it it it , , SSL D t T i WB it is 0 , , 0 WB F i t tat i t s t ssld it it i T , , , , SSL O I U t i WB it it it it it it it it 12
Recommend
More recommend
