When the general stores configuration, supply and distribution networks have been defined, the majority of reorder problems can be solved automatically by means of computer programs, which use generally available statistical information.

This manual aims to provide information about business processes of inventory management, in order to run programs, analysis and reports, calculate parameters and provide logistic operations and decision support. Only after proper setup, can programs, analysis and reports be run. The working of the programs can be modified by constants and tuning factors defined in the setup.

This manual contains the program descriptions related to optimal reordering and stockholding of inventory for production or simulation.

In addition to logistic parameter calculation, the following item statistics are recorded.

First/last customer order date

The total number and quantity of customer orders.

Quantity per issue and cumulative distribution

The average normal quantity per issue and deviation

Average daily issue quantity and deviation

Up/down year trend signal

Seasonal pattern signal

Lumpiness signal

Last 12 months total number of customer orders.

Last 12 months total quantity demanded.

Last 12 months turnover value.

Last 12 months number of backorders.

ABC Classification item by number of customer orders

ABC Classification item by turnover value

ABC Classification item by turnover profit

Exception threshold level

Stock-out dates and periods

Lead time distribution

Average weighted lead time and deviation

ABC Classification suppliers by order frequency and value

ABC Classification customers by order frequency and value

The logistics inventory management system LOGIC aims to provide an optimal logistics of demanded materials by getting

1. " the right item in the requested quantity of the required quality in the right place at the right time "

This problem involves the definition of a network configuration presenting an optimal structure through which the products will flow from their source points to their demand points, and determining what facilities should be used, how the facilities should be served, how many there should be, where they should be located, and which transport services should be used between them. Balancing capital costs, order processing costs, costs of holding stock buffers in stores, cost of non-serving of demand, transportation costs, customer service targets, dictate how the products flow through the network and which facilities, stores need to be used.

The general methods by which the alternative configurations can be evaluated on cost and efficiency is a planning problem, and not part of optimal reordering logistics (configuration and planning). The procedures and programs described hereafter, aim at finding optimal reordering for a fixed configuration.

In order to determine the optimal ordering conditions, quantities and methods, for a given stores configuration, it is very important to distinguish the different types of demand to be served, and different sources of supply.

The definition that demand should be served by "the right item in the requested quantity of the required quality in the right place at the right time, at the lowest costs" leads to a problem that has both spatial and temporal aspects. The spatial aspect refers to the location of the inventory such as warehouse, retail outlet, detail or sub store, or supplier kept depot. The temporal aspect is one of availability of the item through purchase order response time or maintenance of an inventory in stock near the customer or short production / assemblage response time.

If time is not critical, e.g. for planned material requirements (C2, E2) with a planned date far enough in the future to ensure that ordered material will be available in time.

 If the time aspect is very important (C1, E1), stock can be placed as a buffer between the demand and the supplier, in order to serve the demand immediately from the stock.

 Having a stock implies additional stock holding costs (storage, handling, and fixed costs), these are balanced by the immediately availability of the item. An additional factor by which the stock holding can become economic as compared to serving the demand by ordering, even if the time aspect is not important, is given by the ratio of number of demands served from stock, to the number of orders needed to replenish this stock. One purchase order may serve many demands, and only one inspection is required.

In general it is economical to serve the demands (C2, E2) by ordering, unless

However, to ensure that the demand plan date is met, it is recommended that the item is ordered with a desired delivery date sufficiently long before the demand plan date, and be kept in a depot until delivery on the demand date.

In general it is economical to serve demands (C1, E1) from stock, unless

One of the most important costs when keeping an item in the stock inventory, is caused by the possession of the item. The value of stocked items is not available to the company or making any profit, unless the item is served. The quantities hold in the inventory are therefore determined as a balance between service level, possession costs and order costs. The quantities of reordering and holding in stock are based on different criteria for customer consumption and spare parts.

Serving infrequently demanded very large quantities from stock will lead to stock-outs and low service levels. To ensure a normal service level when serving these occasional very large demands, very high safety stock quantities will be needed. It is very economical to separate these exceptional demands from the normal demands served from stock, and serve these exceptional demands by a special order from the supplier, the so-called Direct-Out.

Quicker serving with low handling costs can be achieved by placing sub-stores, local outlets or self-services stores near the production location, or location of material demand. The costs of inter-warehouse transfer are balanced by the lower handling and distribution costs and the ratio of one transfer to number of requests served by this transfer.

Items can be supplied to the store by means of different order methods or policies. The benefits of these different methods lies in the fact, that they lead to lower transport cost by bundling the order-lines for the same supplier or manufacturer, lower the stock value by just in time ordering of the required quantity of a single item, or improving the reliability by ordering on a agreed fixed date, which can be used by the manufacturer in his production schedule.

Keeping a supplier catalog and supply history, allows to actively chose among a large number of supplies and selecting supplies optimal meeting preset requirements

The success of meeting the requirements of customer service can be measured by a number of critical success factors or key performance indicators:

Customer service level is defined as the ratio of the number of fulfilled demands served in time and the total number of demands

Service level of Backorders or Requests that can not be served within the predefined time, leads to unsatisfied demands or Backorders.

The time necessary to serve the backorder is determining the service level of the backorder.

The ABC analysis is used to differentiate items into a limited number of categories, by ranking them according to a given criterion such as number of customer orders, sales, profits, market share or competitiveness, or number of breakdowns when the item is used as spare part.

Usually a small number of items are contributing a high proportion of the sales volume.

This disproportionality between the percentage of items in inventory and the percentage of sales is sometimes referred to as the 80-20 principle, although rarely do exactly 20 percent of the items represents 80 percent of the sales. "A" items are typically the fast movers, "B" items the medium movers, and "C" items the slow movers.

To the standard ABC analysis for ranking 2 categories are added "N" and "D".

The category N presents the recently introduced items, which exist less then a year. The category D presents items with no or zero movement, or turnover for the last 12 months.

The standard used classification in categories by demand frequency or number of customer orders, and turnover value together with their default threshold values are denoted below:

A First 80 % of the cumulated value.

B Next 15 % of the cumulated value.

C Last 5 % of the cumulated value.

N Recently introduced, less then a year.

D Not moving, zero criterion value.

The threshold values for the categories may be changed in setup function INU01/1

After ranking, sample in groups (e.g. total criterion in first 1%, next 1%, etc) and calculate the cumulative distribution.

If all members are equal

one gets after ranking equal criterion values

If the members are NOT equal, one finds after ranking the first 80 % is already reached in the first 5% of all the members.

The ABC analysis can be used for a wide range of purposes: some examples are given below.

Physical grouping off items with a high issue frequency by ABC analysis by demand frequency in locations in the vicinity of the demand areas e.g. the counter in a hardware store, may facilitate and reduce the picking workload and lead to shorter average distances and picking times.

To increase the efficiency of the cycle counting procedure, A-category demand frequency items can be assigned a higher counting frequency.

The careful manual control of each individual item is usually too cumbersome.

Grouping items in broad groups allows for effective and efficient collective analysis.

The overall control of the investment in inventories, the average service levels, can easily be visualised by summary reports over the ABC categories.

The ABC categories can be used managing the problem size. Since the A group dominates the other categories for the given ABC criterion, focusing attention on only items of this group will reduce effort while retaining accuracy in representing the problem.

Combination of the ABC analysis for two or more criteria, can provide even more effective grouping. By presentation of multiple aspects in matrix form, very clear views can be obtained.

By selectively applying inventory policies to the different ABC categories, inventory goals can be achieved with lower inventory levels or higher service levels, than with a single policy applied to all product classes.

The setup function INU01/5 allows to assign weighting factors to influence the safety stock in the various ABC categories.

A combination of the A category customers with the A category of the item issues, focuses on who are the most important customers, and what was their most important demand.

The ABC analysis on order frequency and amount on CUSTOMERS, ITEMS and SUPPLIERS can be used to select by query the particular category one is interested in. e.g. fast movers or non-moving items, or items with a high turnover

This function calculates and displays the results of the ABC analysis over a range of subjects and for user definable cumulative thresholds. For every category the % of the total number of items over which the analysis is run is shown together with the actual number and amount in the category.

(1) Choose the subject, e.g. item classes and a particular item class

(2) One may change the cumulative thresholds

(3) Start the analysis by mouse clicking the start button

(4) One may focus on the A, B area by choosing the first 50 % or 25 % and restart the analysis

A general flow diagram is depicted below.

The programs focus on the different areas

The spare part cost risk analysis is somewhat different since it focus on "having" rather then "re-ordering". Where ROP and FOC parameters are calculated by demand forecast based on historical consumption, reliable historical consumption data of capital spare parts is normally not available.

The function displays the historical network material flows or events by subject, such as customer consumption, transfer and order amounts, sampled in time periods.

e.g. The screen above displays the In/Out balance for the "MS" store sampled by month.

In = received purchase orders (red), Out = transfers to the satellite stores (blue) and issues to the customers (yellow). NOTE: If the "red" > "blue" + "yellow" the stores stock value will go up.

Although the store is mend to serve material requests, non-recurrent requests for exceptional large quantities can disturb the servicing of normal requests, and can be more economically handled by direct ordering instead of servicing from the store. Correct handling of these non-recurrent or infrequent exceptional requests is very important. The penalties in costs and service level can be very high if these exceptional large demands are not correctly identified and handled. Serving a demand for an exceptionally large quantity from stock will generally lead to stock-outs and backorders for the normal requested quantities.

Characteristics of the exceptional large demands are:

Requests for exceptional large quantities can be readily distinguished by their exceptional statistical behavior: they are separate from the bulk of the distribution, where the bulk of the distribution will usually be found within the mean A plus a number of standard deviations D.

To identify exceptional large quantities the criterion is used

Sometimes D is very small or zero (if for a large number of demands always the same quantity is requested) leading to a more logical definition E > A + N x greatest (A, D).

A factor N in the range of 3 to 10 will generally separate the exceptional events from the normal ones: If the requests for material would follow a normal distribution, the possibility of a request of Qty > A + 4 x D will be less then 0.02%

The penalties for not identifying the exceptional demands are very high

Special care has to be undertaken excluding the exceptional large quantities in the calculation: The average and deviation used in the expression above are obtained by an iterative procedure, first calculating the average and deviation from historical requests, and recalculating of the average and deviation excluding the exceptional from the calculation.

Economical constraints can be imposed by shifting the N to a smaller value if the value corresponding to E is very large. This constraint will lead to quicker ordering by direct out of expensive item demands. To avoid a very small value of N, a minimum can be defined.

Economical constraints can be imposed by shifting the N to a larger value if the value corresponding to E is very small. This constraint will lead to nearly always serving from stock of cheap item demands. To avoid a very large value of N, a maximum can be defined.

If there is sufficient quantity in stock, the exceptional large demand may be served from stock. This is the case if after serving at least one reorder point + average daily consumption is left.

In order to avoid checking this condition a so-called stock dynamical exception threshold level can be calculated such that after serving at least

(Stock Dynamic Factor) x (the reorder point + daily consumption)

will be left in stock for normal serving and reordering. This stock dynamic exception threshold level, is dependent on the situation of the stock, therefore it should be regularly recalculated.

The following constraints are applied to the exception threshold level

Forecasting is aimed to predict the future consumption based on the historical consumption. Prediction of future consumption is important for two purposes:

Forecasting can be run

• By user defined fixed method for each item.

• By automatic selection of the method most successfully predicting the recent known history.

The automatic selection is based on the principle that whatever would have worked best in the recent past should be used to predict the future.

A large number of different forecasting strategies is used to forecast the most recent historical periods of demand. Each formula is measured according to variance from what actually happened, and the formula which produced the most accurate forecast of the recent past, is assigned to the item and used to calculate the forecast and deviation of the future.

By dynamically choosing the method giving the best fit to the historical consumption one ensures that the forecast will automatically adjust to sudden changes of the consumption pattern by selecting the best fitting method to accommodate these changes.

The material supply needs for a future period includes a number of terms

The sum of this material need must be balanced by

If the total quantity thus obtained:

(QF1+QF2+AL+SP+SS+SR) - (OH + OO)

is positive, there is a material supply need.

This supply need must be expressed in whole orders, where an order is taken as the greatest of the economical order quantity EOQ and the supplier minimum, and the result is expressed in supplier multiple condition.

see report - material requests projected by supplier

Forecasting predicts the future consumption based on the historical consumption.

Prediction of future on a statistical basis can however be overruled if one has non-statistical knowledge of the future, e.g. there is a non-statistical expected growth known to be happening as consequence of acquisitions or a newly developing consumer market.

The planned quantity or known future need, which has to be used instead of the statistical forecast, can be entered in the functions

INU04 /3 - Stock manager

INU12 - Exceptions manager

To restrict the validity of the planned quantity, an end-of-plan date or expiration date can be defined. This date (if defined) will when expired erase the planned quantity and its own value, setting the planning and date undefined. When the planning is undefined the statistical forecast is used again.

Update of large groups of stock positions with data from an EXTERNAL PLANNING system is possible by special API interface containing the elements

Store code

Item code

Normal plan quantity

End of-normal plan date

Direct-out plan quantity

End of Direct-out plan date

This interface allows at any time for any items group or particular store, to define planned normal demand and planned non-stock demand or direct out, for any valid time window for which the planning is foreseen.

The logistic parameters will be calculated using the planned demand quantities, in conjunction with known historical demand characteristics like average demanded quantity patterns.

If there are different stores, several replenishment configurations are possible:

Each store acts as replenishment center and orders independently from the others stores. If no principle store is defined on the item level, the parameters are kept on the stock record for each store separately.

Two or more stores, possibly on the same site, but not too far apart can be joined in one multiple stores configuration. This can be done by assigning a principal store on the item level. Objectives to do this are:

The motives to construct a multiple stores installation are based on the assumption that the stores have a significant number of stock items in common, so that the safety stock, the total average stock levels and the number of purchase orders can be decreased, and the order quantities per purchase order increased.

In order for a multiple stores configuration to function correctly, regularly a re-assessment of the inventory levels per store has to be done, resulting in a relocation of items to achieve an optimal distribution over the stores.

By assigning a principal store on the item level in function INU03 one assigns this store to act as a master store dictating the replenishment for the total site, whereas the other stores could be considered as 'slaves' for the item.

However, it might be desirable to handle certain items strictly separated per store. This simply can be done by not assigning a principal store on the item level for these items. Allowing to define a principal store offers the flexibility to choose per item whether the inventories should be accumulated and gives the user optimal freedom to direct the replenishment for each stock item independently.

The stock inventories are accumulated for the principal stores item and one safety stock one reorder point and economic order quantity etc. will be calculated for the principal store.

Note that, the maximum stock and minimum stock level constraints should be calculated by summations of the corresponding individual stores constraints. The re-order policy of the principle store will be followed.

Satellite stores are replenished by transfer from the master store.

NOTE: This configuration can not be considered as a multiple stores configuration, since the transfer is general in one-way. If the inter-stores transfer is not two-way, a large quantity held in a satellite store, would on multi-stores cumulating, block re-ordering at the main store.

Three general classes of costs are important to inventory management. Order or procurement costs, holding or carrying costs, and stock-out or absence costs. These costs are in conflict, or trade-off with each other. The optimal order quantity is calculated as a balance of these costs. The choice of inventory policy can be based on the various costs.

This cost represents the cost of stocking and keeping items. It is normally calculated as a percentage of the item average unit cost of per year, where the same percentage is taken for all items in the warehouse. Having an inventory can cost between 20 and 40 percent of the stock value per year. The cost related to keeping the warehouse and materials can be divided into:

Cost of order or procurement costs. These costs are related to processing the order and ordering an item by order-line. The cost includes cost of reviewing the proposal, creating, approving, printing and sending the order-line, invoice matching, transport cost, receiving, inspecting and entering cost. Costs associated with acquisition of goods for replenishment of inventories highly influence the reorder quantities. When the order quantity is decreased, the total stock value will be decreased, with a (hidden) higher cost resulting from a higher frequency of ordering. The cost related to the cost of ordering can be divided into:

Cost of absence or out-of-stock costs are incurred when a customer order is placed but cannot be issued from inventory. There are several kinds of stock-out costs, depending on the role of the item as consumable or spare part.

A number of parameters are needed in the logistical calculations such as service level, holding cost, fixed and variable order costs, absence cost and review/entry time. The values of these control parameters can be specified on a number of levels. The levels define a hierarchy, which is followed when the value of a parameter is used in a calculation.

The definition hierarchies for the different parameters:

Transfer channels

The hierarchy allows simple data definition, analysis, and maintenance. If for a particular store the fixed costs are different, only define the parameter for this store. If the holding cost is higher for certain items e.g. because they need a special dust free, cooled location, just define and link a parameter group for these items. If the holding cost is higher for all items of a particular class: then do not define the values for all these items, but just assign a parameter group to these items. The definition of the variable order cost allows adjustment of the economic order quantity EOQ to reflect the changes in most economical ordering conditions. E.g. when ordering from a nearby supplier, having low transport costs, the variable order cost will be low resulting in small EOQ quantities with a high order frequency. Change to a supplier located far away and the high variable order (transport) cost will automatically lead to a large EOQ, thereby lowering the overall costs, by lowering the order frequency.

The inventory levels can be controlled by different control policies or order policies.

Which inventory control policy should be chosen depends on the item characteristics, its use, the consumption, price, availability at the supplier, transport costs and various other aspects.

The following inventory control or order policies are supported:

Statistical calculations need to be performed, to support the ROP and FOC and CSP.

The other policies depend on user definition of the minimum level, lot quantity and maximum stock.

The reorder point method, is an economical, very simple method to reorder items. It focuses on the reordering of each item individually, by reordering a (constraint) economical order quantity whenever the quantity in stock becomes equal or drops below the reorder point.

This order point (ROP) is a stock quantity that represents a moment in time at which a reorder proposition should be created. It is calculated in such way that when the material is ordered there should be enough stock during delivery lead time to cover the "normal" demands for material to a certain pre-defined desired service level, before the ordered replenishment is received. In this way it determines the optimum moment in time so as to ensure service of "normal" demands in the expected lead time for delivery.

The order point is calculated using the following information

The order point OP is split in the components stock reserve (SR) and stock security (SS)

Where

The value of SR is obtained straightforward from the expected demand and lead time.

The calculation of the safety stock quantity to assure a certain predefined desired service level, requires a more complex mathematical treatment.

The quantity ordered should reflect a balance between the cost of possession of the stocked quantity of the item and the cost of ordering of the item with the aim of minimizing the total stock management cost.

The optimum quantity to be replenished based on the lowest total relevant cost when the available quantity just reaches the order point OP is given by

The cost of order may include a fixed administrative component or order cost, and a variable component or order-line cost, depending on the transport, handling and inspection, and a component depending on the stockout costs.

The EOQ, minimizing the total cost, will for the same turn-over of an inexpensive item, lead to low frequency ordering of large amounts of the item, and high frequency ordering of small amounts of an expensive item.

The EOQ is only optimal in economical, but NOT in a logical sense, and is adjusted conform a number of rules e.g. Order not more then one years usage to restrict overstock when the consumption pattern will change, order not more then expected to be consumed in shelf-life, order not more then the maximum stockable quantity, etc.

ROP method without orderline consolidation

• Each stock item is treated as a separate management task

• By single item most economical

• Reduce investments in stock

• Different Supplier choice possible

• Direct Out for Exceptional quantities offer NO Problem

• No special measures (Purchase scheduling, etc.) needed

• No purchasing advantages by synchronization items for the same supplier

• Possible problems when supplier = manufacturer with "Out of the blue" orders

Some items have a long manufacturing time, the manufacturing is planned in the manufacturing schedule of the supplier.

The moment of reordering, which is triggered by consumption and is optimal for the stock keeping, may not be fitting in the supplier manufacturing schedule.

The consolidation process aims to synchronize ordering of items for the same supplier and contract, and results in orders with many order lines. It works by pulling other items into reordering, when one item is normally to be reordered for a given supplier and contract. In order to get an item into reordering, the reorder point is artificially increased. A number of limits to the increase of reorder point can be entered.

When one item is normally to be reordered because the quantity on stock OH drops below the reorder point ROP, other items for the same supplier and contract can be "pulled" into reordering increasing the reorder point is artificially.

• Less order cost per orderline

• Less work

• Lead-time is reduced

• Less errors

A number of factors to increase of reorder point can be defined, each limit having its own benefits. A combination of different factors will increase the performance of the method.

e.g.

(1) variation of the reorder point within a certain value, will only bundle low costing lines.

(2) variation of the reorder point within a number of days before the order point will normally be reached, will also bundle high costing lines, which otherwise soon will be reordered.

(3) Combination of (1) and (2) has the advantages of both.

Grouping of replenishment requests for items of the same supplier and ordering on predefined dates will result in the following advantages:

In order to fulfil the requirements above, requests for the same supplier have to be synchronized and ordered at FIXED INTERVAL with a VARIABLE QUANTITY

The fixed interval, cycle period, or review time can be derived in a number of ways:

An economic order period based on the EOQ is calculated for every item, as consumption time of the EOQ. The resulting periods are rounded to 0.5 - 1 - 2 - 3 - 6 - 12 months. In this way the quantity expected to be consumed in these rounded periods will be an approximation of the EOQ and the items can be grouped together by supplier and the fixed order dates can be synchronized.

An optimal common review time can be calculated for jointly ordered items. The resulting period is rounded to whole months.

User defined Any user defined cycle can be assigned to an item.

The variable quantity or maximum level M can be calculated once the FOC cycle period has been defined.

An ordering schedule can then be constructed for the groups of items in order to get a smooth workload spread over the whole year.

• Item purchases are bundled for the same supplier, orders can be expected on agreed dates.

• Lower Purchasing cost due to lower transport costs

• Possible lower lead-times, and lower purchasing prices

• Better for Supplier: Orders for manufactured items on agreed dates

The reorder point method, is an economical, very simple method to reorder items. It focuses on the reordering of each item individually, by re-ordering constraint (supplier conditions, maximum stock, shelflife, user defined lot qty, etc.) economical order quantity OQ whenever the quantity in stock becomes equal or drops below the reorder point.

This re-order point ROP is a stock quantity that represents a moment in time at which a reorder proposition should be created. It is calculated in such way that when the material is ordered there should be enough stock during delivery lead time LT to cover the "normal" demands for material to a certain pre-defined desired service level, before the ordered replenishment is received. In this way it determines the optimum moment in time so as to ensure service of "normal" demands in the expected lead time for delivery.

The SS term is a dominant term in the average expected stock value. The SS is a direct function of the lead time and the desired service level. SS is comparatively small if the lead time is short.

Stock Value = 0.5 x Sum OQ + Sum SS( Service Level | lead time )

May easily overcome the fact that normal ROP - policy does not bundle order-lines for the same supplier. For a supplier of 100 items it leads to regular orders of 20 to 50 lines per order.

The FOC-policy aims at ordering at fixed dates having regular intervals or cycle periods T between these dates. The fixed dates, allow grouping of order lines for the same supplier together with the purpose that the lower cost of transport will outweigh the higher inventory cost caused by re-ordering the item at relatively high inventory levels.

Where the ROP-policy the ROP aims at only to serve the demand in the lead time with a predefined service level, the FOC-policy quantity must serve demand in the whole FOC interval or cycle period T with a predefined service level.

To meet a service level a safety term SS must be added for the whole cycle period T.

Stock Value = 0.5 x Sum CQ + Sum SS( Service Level | Cycle period )

Where the stock-out probability of a single line in FOC re-ordering may be small, probability on having a stock-out of any of the items of an ensemble is rapidly increasing with the seize of the ensemble.

It can be seen that replenishment of stock by FOC policy for large numbers of lines, the single lines must have very high individual predefined service levels to avoid stock-out occurring for any of the lines.

The min-max system is a simple and popular method of inventory control. The maximum level is often determined by constraints, such as consumption in shelflife, or defined by space and stocking conditions, or the economical transfer quantity, or otherwise determined re-supply quantity. When quantity on stock drops below the minimum level, the reorder quantity will refill the stock to a defined maximum level.

This method is frequently used to refill the shelves in retail stores, supermarkets and sub-stores. If there are no additional constraints the maximum level may be set to the reorder point quantity plus the economic order quantity.

The order quantity is not always the same because the amount that the quantity drops below the reorder point is added to the difference between the minimum and the maximum level (refill or transfer quantity).

Since the maximum level is usually not recalculated but remains fixed over an extended period of time, growing demand will result in higher frequency of inter-warehouse transfers and refills. This behavior is opposed to the ROP method where a growing demand will lead to larger economical order quantities.

The low economical cost of refill in time and workload of inter-warehouse transfers and the constant availability of items in the providing main stores, often makes the min-max order policy the preferred method to keep inventory in sub-stores or satellite stores and retail stores.

A variation of the min-max method may be used to manage spare parts or other slow moving items. Some inventory managers consider that for these slow moving items, one should replace stock only as it is used on a 'one for one ' basis, where the maximum level is set to a safety level plus the usual set size or statistical normal demanded quantity of a single demand. On issue of an item, an order is placed to replace the issued item.

Spare part consumption can be contributed to 2 constituents:

Spare parts provide insurance to cover the risks and costs of production stop, when the production equipment breaks down. It may not be economical to hold substantial quantities of all possible spare parts in the inventory. The cost of having a spare part in the inventory is balanced by the cost of not having, and the budget for spare part inventory.

MAX method to manage spare parts.

The stock is replaced as it is issued on a 'one for one ' basis, where the maximum level is set to a safety level plus the usual set size / statistical normal demanded quantity for one demand. On issue of an item, an order is placed to replace the issued quantity.

If not all possible parts can be kept in the inventory, a spare part cost-risk effectiveness analysis may be run, to list which items if hold in the inventory provide the least risk, for the available budget.

No program, however sophisticated can produce better results then contained in the quality of the input data. One simple data entry error may completely vitiate the statistical results. A data error of fools filter is therefore an absolutely necessary in many programs: e.g. The human eye will recognize the data of a receipt in the year 2096 or in 1895 as wrong, however a program will not exclude these data if not programmed! Data error filters are therefore an essential component in all programs.

The optimal value of the calculated parameters is normally obtained in a step by step process where optimal results can only be obtained after having passed several criteria, each representing a different aspect: Statistical, Economical, Logical, and sometimes Dynamic. Only by applying the diverse criteria in a succession of steps will lead to a solution which is most optimal in all senses.

Therefore the solution will generally consist of a function where the optimal value is restricted by boundary conditions. A number of examples are given below.

An exception threshold level, above which a requested quantity will not be served from stock but as special direct-out order, can be calculated, based on the statistical requested quantities distribution, such that only a small well-defined number of the requests will be served by direct-out order. This exception threshold level however does not represent the most economical or logical situation.

An economical order quantity can be calculated based and the various costs and the statistically forecasted consumption. This quantity is however only an indication of the quantity which will lead to the lowest total cost, but not necessary the best solution to a number of other criteria, e.g. the economic order quantity may be much larger then what can be consumed in the shelflife, and when ordered leading to waste by deteriorated products.

An economical order quantity for an item with a limited shelf-life, may be larger the expected consumption in the shelf-life, thus economical is not always logical.

The optimal value of the exceptional quantity may take the actual stock state into account. Since the quantity in stock may vary, this dynamic exception threshold will vary likewise.

Although the programs are designed to be run automatically and calculate optimal parameters which control the logistics of the inventory management, none of the methods or parameters "takes the control of your system over", the parameters just provide optimal values in a statistical sense. The product responsible keeps full control, by the ability to enter his or her own preferred parameters, or preferred forecast method by use of the screens INU04 and INU012 if he wishes to do so. If no user definitions are provided the statistics is used.

A program can run automatically in batch mode, or be started from the menu.

To start a program from the menu, one first has to select the PROGRAMS menu.

To start a program, click with the mouse on the program icon.

The supplier statistics and items lead times program calculates the statistics of the delivery delays, the numbers and total amounts, of ordered received materials.

The following statistics are calculated by supplier and supplied items. The calculation can be weighted to focus on recent history. The calculation uses orders for re-stocking as well as direct out orders.

1. Lead time and deviation for all receipts for the time window and weighting scheme and conditions as defined in INU01 /6

2. Purchase history and ABC analysis in number and total amount of ordered for the time window and percentages as defined in INU01 /1

Results of calculation on the tables - SUPPLIERS, CATALOG

See function INU06 /1 and INU06 /2

The items statistics program calculates general single demand characteristics

The program performs 2 steps:

1. Average expected normal demand and deviation for all customer orders for the time window and weighting scheme and conditions defined in INU01/7

2. Exceptional quantity threshold or break quantity using the conditions defined in INU01/7

Results of calculation on the item stock positions table - STOCK

See function INU04 /2

NOTE: The single demand statistics presents a very important factor in the safety stock calculation for the ROP and FOC reorder policies. Service levels and stock values are dependent on the accuracy of the single demand statistics calculation.

The items statistics program calculates general consumption statistics

The program performs 2 steps:

1. The selection of number and total amount of the issues, returns, direct-outs, transfers and served backorders for the time window, and ranking by ABC analysis of the total turnover number, amount (quantity x sales price) and gross benefit (quantity x (sales price - cost price)) over all stock positions. Consolidation over all stock positions (different stores where item is held) to the items level, ABC analysis of the total turnover number, amount and benefit over all items.

2. First and last transactions date, total number and quantity all customer orders

Results of calculation on the item stock positions table - STOCK

See function INU04 /2

NOTE: The various ABC analyses over the items and stock positions provide very important management information.

The customers sales program calculates the general consumption statistics for the customers, organization units, divisions or companies.

The statistics are calculated per customer. The calculation uses all customer orders, for stock items as well as direct out orders and the time window and ABC percentages defined in INU01/1

The program performs the calculation step:

The selection of number and total amount of the issues, returns, direct-outs and served backorders for the time window, and ranking by ABC analysis of the total turnover number and amount (quantity x sales price) for all customers.

Results of calculation on the customers table - CLIENTS

See function INU07 - Customers

This program calculates the short and long forecasted consumption for all the stock positions, the normal expected consumption as well as the expected direct out consumption.

The forecasting is for re-ordering conditions, order quantities and contracts and described in the chapter

The forecasting setup, forecasting formulas and methods are described in the SETUP manual function MAU01.

The use of the forecasting in the calculations is conditional and can be overruled by planning as described in the chapter

The forecasting program also sets a number of stock consumption characteristics

Results of calculation on the item stock positions table - STOCK

See function INU04 /3

The forecast for each stock position for normal and direct out demand is displayed on function INU04 page 3, on the same page a user defined planning can be entered, with an expiration date.

The program calculates various target and steering parameters obtained by consolidation of the target parameter hierarchy as the real costs and targets for each stock item position.

These cost and target parameters are necessary as input for the reorder parameter calculations performed by INX07 and INX08 See SETUP manual and chapter

The program also calculates a time smoothed average stock quantity, and keeps a record of the stock-out periods for management information.

Results of calculation on the table - STOCK

See INU04 /1

The costs and times for the same item in two different stores:

Main store with default supplier (TELI50)

Satellite store with default supplying main stores (MS)

The program calculates the re-order parameters for the ROP policy.

It uses the costs and targets calculated by program INX06 and shown by function INU04/1

The results can be affected by constraints as defined in INU01/4, INU01/8 and INU03

The orderline consolidation is also part of the program.

The results by stock position are displayed in function INU04/4

The economic order quantity EOQ as function of expected consumption, cost price and various costs such as cost of order, cost of holding and cost of stockout, is only optimal in economical sense, but not in a logical sense, and can be corrected by a number of rules

To restrict overstock when the consumption pattern will change:

To order not more then can be held in stock, or consumed in the shelf-life

To restrict the number of orders per year, and the order frequency per year

To restrict the number of concurrent orders when the lead times are long and/or the demand rates are high:

Note: a number of these constraints may be in conflict, e.g. The minimum order quantity of the supplier, versus the maximum stockable quantity. A long lead time, versus a short shelflife.

The program re-calculates the various order parameters of the FOC policy

The FOC policy aims to order all items for the same supplier together, by reviewing the stock positions on regular fixed dates with constant periods of time between them.

It uses the costs and targets calculated by program INX06 and shown by function INU04/1

The calculation can be affected by constraints as defined in INU01/4, INU01/8 and INU03

For more information see chapter

For a discussion of the benefits of using the FOC versus the ROP policy see chapter

The results of the calculation by stock position are displayed in function INU04/4

The FOC scheduler INU11 can be used to initialize the different parameters needed by the FOC policy, such as

Maximum level The quantity ordered will refill the stock position to this level

Interval Interval between stock reviews

The purpose of this function is to select the stock items for a particular supplier and contract and calculate the interval I, maximum level M and trigger level T by use of a selected FOC scheme. After the first review date D is defined, the policy can be changed to FOC-policy.

The list and go buttons allow to

1. If the field value is not defined, select a valid value from a list
2. If the field value is defined, start the step

The clear button resets the FOC definition back to ROP and erases defined FOC parameters