BOM structure method
Material list BOM is one of the three basic input information running MRP. Different structures can reduce basic data preparation effort, improve the speed of the material decomposition algorithm. This paper analyzes the construction method of several Chinese BOM in detail for reference.
First, conventional methods
In the traditional MRP software, the BOM is a network-like structure stored data, so there may be a parent item number below with many sub-numbered numbers, and a child number is the case of different parent numbers. The data structure is: parent number, child number, structure, low layer code. The software can indicate the hierarchical relationship between the material by the project indent.
Advantages: 1) Good adaptability, small storage data.
2) Evenable material decomposition and material summary.
Disadvantages: 1) Need to prepare a large number of programs to achieve intuitive display products.
2) Not easy to perform the case where the retro parts are suitable for the product.
Second, the tree structure method
In the MRP software, especially in the MRP software under the Windows platform, typically BOM uses a tree-type structure, and its data structure is: Treekey, Parent, material coding, quantity, decomposition marker. Where the treekey identifies the node number, the Parent identifies the parent node.
Advantages: 1) Use the TreeView control of the Windows platform to achieve the addition, deletion, copying of the node.
2) The interface is beautiful, intuitive and easy to understand, and the user is easy to operate.
3) Construction of the product BOM in a single piece of small mass production method.
Disadvantages: 1) For many series of multi-product, the amount of data will expand sharply.
2) Not easy to perform the case where the retro parts are suitable for the product.
3) The material decomposition algorithm is complicated, and the improper handling efficiency will be low.
Third, the logo method
This method is suitable for multi-series multi-product, and each different parts are identified, and their suitable series and product models are used in the corresponding model marker position.
For example, a locomotive factory has 17 product lines, each series of approximately 20 to 25 different models, each product has 80 ~ 90 parts, using treekey, Parent construction BOM, its number of records is approximately 17 * 20 * 80 = 27, 200, the amount of data is very large. Edit, modify, calculate BOM possible efficiency is low.
The symbol method can be used to construct BOM according to each series of buildings. Problems: The number of parts per range is approximately 150 to 200, and the total number of records is approximately 2550.
1) Construction method:
· BOM table structure: materials coding, structure, series, adapt model, material category
Note: When the material adapts to a model of the series, its logo location 1
Material categories are divided into: finished products, self-cultivated, external purchases, external access
· MARK table structure: series number, material code, code bits
Note: The model flag corresponding to each product is stored here.
2) Material decomposition algorithm
Remove one record in the MPS, query bom.dbf, if the material is a finished product, query mark.dbf, remove the series number-> XLBH, code bits -> MW, remove the series of BOM.DBF = XLBH, adapt The record of MW = 1 in the model.
The number of demand demand in MPS is * the number of structures in the BOM is both parts. This algorithm does not use recursive, a simple SELECT statement, high efficiency.
Advantages: 1) Suitable for multi-series multi-product, basic data can be sorted by series, implement small workload. 2) The material decomposition algorithm is high, and the speed is fast.
3) Easy series and model suitable for auto-check parts.
Disadvantages: 1) Requirements BOM only one layer, the series is root, the part under the series is the leaves, and the adaptability is limited.
2) The composition of each product cannot be displayed in an intuitive manner.
Fourth, modular BOM structure
Modular BO is mainly used in multi-series. This product consists of a basic piece, a feature, an option, wherein a variety of features (must be one) thus can constitute different products. For example: truck production plants, there are 10 kinds of engines, 2 types, 4 chassis, 30 colors, can form 10 * 2 * 4 * 30 = 2400 products, if stored according to product structure, it is necessary to deposit 2400 Structure and make MRP material decomposition is complicated. The modular BOM configuration is used, remove the product layer, and the component layer is used as the final state, and its structure is only: 10 2 4 30 species. The amount of data is greatly reduced.
The BOM table data structure is: parent number, child number, option number, number of structures, prediction ratios
Note: Option number representation: basic parts, feature serial number, optional
MRP material decomposition basic algorithm: according to the number of demand requirements of MPS, decomposes the number of basic parts and features (quantity * Structure * Structure * predictive ratio).
