Spare Parts Planning For New Metallurgical Plant Start-Ups—Why You Need A Spare Parts Inventory Model

Printable Version (PDF)



Standard practice for newly constructed metallurgical plants is for equipment suppliers to recommend sufficient spare parts for the first two years of plant operation. However, this approach is inadequate, as it does not consider several factors, including:

All of these questions are critical and are not adequately answered if the equipment suppliers simply deliver an estimated two-year supply of parts.


All too often, operations preparing for a new plant start-up limit spare parts inventory requirements to equipment suppliers’ suggested spares—usually the recommended first two years’ spares. The end result is a collection of disparate printouts in various formats. In most cases, the engineering company simply orders the spares, which are thrown into a lay-down yard or into a shipping container. Is this useful for the plant owner/operator? The simple answer is no; it is not very useful.

What the plant operators really need are data that can be uploaded into their purchasing and warehouse inventory control system. For each part, these data include:

Of course, these values are not determined directly. They are computed based on answers to the following questions:

With the answers to these questions, along with values for the cost of carrying inventory and preparing purchase orders, the key parameters—ROP and EOQ—can be automatically determined by an inventory model.

Developing a spare parts inventory model, requires input from engineers and plant operations specialists with years of experience in the operation and maintenance of similar metallurgical process plants. This experience can be put to use in determining the necessary insurance and routine spares, as well as estimated usage rates and recommended safety stock requirements.

The parts data come from a variety of sources, including:


Each equipment supplier is contacted by e-mail or telephone, and a recommended spare parts list is confirmed. The supplier-provided spare parts recommendations are then scrutinized by experienced personnel. Part type and usage rates are determined by these personnel based on plant experience. The parts analysis team must capture or establish the following data for each part:

Part Number

The manufacturers’ part number is entered in this field.

Part Type and Description

Using a consistent naming convention, the most descriptive terms are entered for the part in the part type and part description fields. In many cases the supplier provides poor part descriptions. Experienced owner or consultant personnel can improve upon these descriptions where necessary.

Part Weight

The equipment supplier-provided weight is used.

Part Manufacturer

When made available by the supplier, the part manufacturer is entered into the spare parts inventory model database. If the supplier does not provide the manufacturer, the supplier is entered in the manufacturer field.

Part Supplier

The name entered in this field is the supplier who has been issued the original purchase order for the equipment.

Unit of Measure

Normally the unit of measure is each. In some cases, such as conveyor belting and wire rope, the unit of measure could be feet, yards, or meters.

Spare Type

The spare type is either insurance or routine. An insurance spare represents a part that may never be needed, but must be kept in inventory in case it is needed. Normally insurance spares are expensive, critical, and have long lead times if one is to be ordered. If an insurance spare is not available, extensive production losses result. Where the spare type is defined as insurance, the ROP is automatically set at zero with a reorder quantity of one.

Routine spares are estimated to be needed in the course of operation. For routine spares, the normal EOQ and ROP algorithms are used.

Lead Time

The lead time is based on estimates of the following:


The total price for the part is entered into the database. If applicable, factors can be added for customs duties and shipping costs.


A component name is added to this field when it is applicable.

Issue Quantity and Annual Usage

The values must be carefully assessed based on equipment use, plant environment, and the experience of the personnel making the analysis.


A linking table in a database is necessary for identifying each equipment item number on which the part is used. For example, there could be several pumps of the same design, each with a distinct equipment number, that use the part in question.

Safety Stock

Safety stock represents a buffer quantity of a part kept in stock. This buffer is designed to account for uncertainty in the lead time and parts usage. In an ideal world, the exact parts usage rate in a specific period of time—say one month—would be known. In the same ideal world, the exact lead time to obtain the part would be known. In that case the reorder point would equal lead-time usage (LTU). For example, if a quantity of one particular part was used each month and the lead time was two months, the reorder quantity would equal two, since two parts would be used during the two months it took to order the part and get it into stock.

Safety stock values are typically based on a multiple of LTU depending on the criticality of the part. For example, consider the following example rules for establishing safety stock:

Safety Stock = 6 x LTU.

Safety Stock = 4 x LTU.

Safety Stock = 2 x LTU.

The higher the safety stock value, the lower the probability of a stock-out for the part. However, the higher the safety stock, the higher the inventory investment.

Other Factors

Other factors necessary to consider before calculating EOQ include:

Cost of Carrying Inventory Expressed as a Decimal

This cost includes:

All of these costs can be expressed as a percentage or decimal based on the cost of borrowing money to cover them. This cost largely depends on the prevailing interest rate. For purposes of this paper, we can assume that it is 15 percent, or 0.15.

Cost to Place a Purchase Order for the Part

This cost includes:

This estimate does not have to be particularly precise. For purposes of calculation, we assume that it is $100.00. Of course the ordering cost does not generally depend on the cost of the part ordered.


Use the following equations for ROP and EOQ:

(1) Reorder Point (ROP) = LTU + Safety Stock

Figure 1 illustrates the concept of calculating reorder points based on the ROP = LTU + Safety Stock formula.


Calculating Reorder Points

(2) Economic Order Quantity (EOQ) = [2BA/I]1/2 / Calculating Reorder Points


B = Cost to place a purchase order.
A = Annual usage cost = annual usage x cost of part.
I = Cost of carrying inventory expressed as a decimal.
Q = Quantity.

The advantage of the EOQ approach is that it minimizes the sum of the ordering cost and the inventory carrying cost. Refer to Figure 2.


Minimum Inventory Cost Curve


The data base information collected, established, and calculated are then uploaded into the computerized inventory management system. Once this process is completed, plant operators and maintenance personnel have an inventory of spare parts and a system of reordering to ensure that parts are available when needed.

To recap, the essential factors that ensure a good inventory of spare parts on hand are:

During actual plant operation, the part usage values can be adjusted based on experience and the ROP and EOQ values recalculated.