Equipment failures can lead to production downtimes, which can cost businesses millions of dollars in losses. Having spare parts readily available reduces downtime — but enterprise procurement teams often overstock their spares inventory, which expends capital and increases inventory management demands. There’s got to be a way to balance “too much” and “not enough” — isn’t there?

There is, now. This new GEP white paper, “A New Approach to Spare Parts Inventory Management,” outlines a scientific method – based on reliability engineering principles – for optimizing inventory and minimizing downtime. It also shares strategies for dealing with excess inventory and managing the implementation of the new policy.

Read this report to gain insights into how procurement teams can run their inventory management operation like a well-oiled piece of machinery.

What’s Inside

  • Applying reliability engineering principles to calculate optimal spares inventory
  • Practical suggestions for dealing with excess inventory
  • Change management and governance methodologies


Availability of spare parts is critical to ensure continuous and reliable plant operations. “Always available whenever needed” is the motto of maintenance teams and plant operations personnel. While this practice is logical, it often results in excess inventory — with studies indicating that almost 40 percent of spare parts inventory remains untouched for at least four years. This not only blocks the capital that could have been invested to fuel the company’s growth, but also consumes space and is an additional cost to manage.

This white paper shares the best practices in spare parts inventory management used by market- leading enterprises. It discusses strategies to optimize inventory, reduce excess inventory and implement continuous governance policies. The paper takes the conversation further by sharing the common challenges encountered when implementing these new approaches and discusses mitigation strategies.

Since industrialization, several methodologies have been successively used to maintain equipment and to improve uptime. These have included: repair after failure; inspecting and performing maintenance as needed; periodic preventive maintenance; predictive maintenance; and cognitive computing (see Fig. 1).

Figure 1: Maintenance Management Strategies Source: University of Tennessee at Knoxville (adapted)

To accelerate maintenance activities, every large manufacturing setup maintains an inventory of spares comprising items that are most critical to manufacturing operations or prone to failures. The number of spares maintained in inventory for any particular item has traditionally been defined by perception — without any established methodology to arrive at the number. This often results in a huge inventory of spares that blocks significant capital and increases costs associated with managing inventory. Now, enterprises can calculate the minimum inventory to decide how much is too much and accordingly optimize it.

However, optimizing spare parts inventory isn’t a straightforward activity — it often demands a thoughtfully designed, multi-phase approach — and cannot be achieved overnight. This paper defines a scientific plan to improve spare parts inventory management and optimize stocks — freeing up significant working capital and reducing operational costs in the process.


Assembling a core team, with representatives from key functions — such as procurement, supply chain, engineering, and IT — based on the nature and domain of business is the most important step to achieve effective inventory optimization.

This team should define the guiding principles, desired outcomes, and risks and timelines,and communicate them to all stakeholders. It should describe the methodology being adopted and support it with objective data points, not subjective constraints. Also, given the breadth of the project, it’s vital to have buy- in at all levels.

If communicated clearly, no enterprise will back away from benefits such as a cleaner, more organized storeroom, reduced carrying costs, money back to the organization via disposal and/or resale, fewer parts to cycle count, and so forth.

As part of the change management process, the core team should determine the forum, frequency, and communication messaging at the onset of the program. The team should also involve key decision-makers and change advocates to identify challenges and address any concerns.


Framework for Reliability Calculations

Reliability can be loosely defined as the probability that an item will continue to perform its intended function without failure for a specified period under stated conditions. Quantitatively, reliability is the probability of success. There are two scenarios that demand attention when developing the framework:

  1. Spare parts for new equipment where there is no consumption history
  2. Existing spare parts where there is consumption history

Spare Parts for New Equipment

Reliability specialists often describe the lifetime of a population of products using a graphical representation called the bathtub curve. The bathtub curve consists of three periods: an infant mortality period with a decreasing failure rate, followed by a normal life period (also known as “useful life”) with a low, relatively constant failure rate, and concluding with a wear-out period that exhibits an increasing failure rate.

Figure 2: Bathtub Curve

The bathtub curve does not depict the failure rate of a single item but describes the relative failure rate of an entire population of products over time. As seen in Figure 2, Zone 1 is the infant mortality period and is characterized by an initially high failure rate. This is usually attributed to poor design, substandard components and/or inadequate controls in the manufacturing process. Generally, a piece of equipment is released for actual use only after it has successfully passed the “burn in” period. A 48-hour “burn in” is usually adequate to significantly reduce infant mortality failures, and so it’s good practice to include the “burn in” run as part of the contract with manufacturers.

Zone 2, the useful life period, is essentially characterized by a failure rate that is constant and resulting from strictly random or “chance” causes. These “chance” failures occur when the stress levels on the equipment/part during operations exceed the maximum threshold due to random, unforeseen or unknown events.

While reliability theory and practice relates to all three types of failures, its primary area of focus relates to chance failures because these occur during the useful life period of the equipment.

Although these failures occur at a constant average rate, the number of events occurring in any time interval is independent of the number of events occurring in any other time interval, and are closely associated with exponential distribution.

Zone 3, the wear-out period, is characterized by an increasing failure rate attributable to equipment deterioration caused by age or usage. The only way to prevent these failures is to replace or repair the deteriorating component before it fails.

While Zone 1 failures can be avoided and Zone 3 failures are experienced in equipment at the outer limits of its usable life period, our focus is on Zone 2 — where the failure occurs during the useful life period of the equipment. Data pertaining to the mean time between failures (MTBF), total number of units being operated at a location, and the lead time of the component is collated and assessed mathematically to derive the expected number of failures. Using this, the number of spare parts required at different service level is calculated.

Figure 3: Screenshot from GEP Spares Inventory Management Tool Source: GEP

Spare Parts for Existing Equipment

Spare parts available in the inventory often have a consumption history — data that can be leveraged to determine the necessary safety stock. While having five years of data is ideal, the analysis can be performed with a minimum of three years. Using this data, SKUs can be segmented based on their consumption patterns into:

  • Very Low Consumption: Less than one unit/year
  • Low Consumption: 1-299 units/year
  • Mass Consumption: 300+ units/year

The safety stock for each segment has to be calculated using different methodologies. Before we perform the analysis, it’s desirable to not consider the items that were recently created — say, three years before the date of analysis — as the probability of consumption of those items is minimal.

[H5] Very Low Consumption: Whenever the consumption is less than one unit per year — or zero — the decision should be made on whether the part should be in stock or non-stock. The core team should perform a vital, essential and desirable (VED) analysis at SKU level considering the following seven major parameters:

  • Functionality: Effect of the component failure on the system’s availability
  • Response time: Duration between call logged to restoration of component’s functionality as agreed in the contract
  • Lead time: Duration between placing the order and the delivery
  • Nature of the component: Commodity/OEM/Fabricated
  • Lifecycle: Which of these phases — introduction, established to be continued, or phase-out
  • Demand: Overall consumption pattern
  • Price: ABC analysis based on unit value

These parameters are evaluated with inputs from the core team using Analytic Hierarchy Process (AHP), and the individual weighing factor is determined at a category level such as bearings, motors, pumps, etc. VED scoring is determined at an item level after assessing the seven parameters — some at a category level (functionality, response time, and lifecycle) and the others at an individual level.

The sum product of the overall weighing factor and the major factors score provides the total VED score for each SKU. Depending on the score, the parts are classified further as Vital (score >4), Essential (score 3-4) and Desirable (score <3). It’s recommended to have vital items in stock and the rest as non-stock items.

Low Consumption: By definition, the consumption is between 1-299 units per year. The major inputs are five- or three-year consumption on a daily basis, maximum demand on a day, and the lead time. Based on the consumption data, the mean and variance are calculated. If the mean times 10 percent is greater than variance, Poisson distribution analysis is performed; otherwise the Bootstrapping methodology is implemented using consumption data from five years to determine the safety stock.

Safety Stock Calculations

If Mean x 1.1 > Variance → Poisson distribution

If Mean x 1.1 < Variance → Bootstrapping methodology

Mass Consumption: By definition, in this case the consumption is 300 or more units per year and they represent normal distribution. The formula to calculate safety stock for mass consumed items is:


α is the service level and Zα is the inverse distribution function of a standard normal distribution with cumulative probability α

E (L) and σL are the mean and standard deviation of lead time

E (D) and σD are the standard deviation of demand in each unit time period

In a Snapshot

Type of Item Definition Methodology
Very low consumption

Less than 1 unit/year

  • Decision to keep 0 or 1 unit in stock
  • VED analysis using AHP procedure — considers: Functionality, Response Time, Lead Time, Commodity/OEM, Lifecycle, Demand, and Price.
  • Vital spares: 1 unit in stock; Essential and Desirable spares: 0 unit in stock
Low consumption

1-299 units/year

  • Apply Poisson distribution or Bootstrapping methodology to determine the safety stock at a desired service level
Mass consumption

300+ units/year

  • Normal distribution (consumption is symmetrical)


For organizations that can identify excess inventory based on the segmentation and data profiling process, the next logical step is to develop a sustainable strategy to remove the excess. After performing the analysis, most organizations are able to identify opportunities for reducing their inventory value by 20–45 percent, on average. Obsolete parts, parts associated with a retired asset, damaged items, and addressable inventory exceeding the “calculated maximum” are all part of this opportunity. This excess inventory can be addressed/avoided by the following means:

Supplier Buy-Backs

The first step in reducing excess inventory is to consider supplier partnerships. In some supply agreements, terms surrounding buy-backs are included and often looked over by procurement specialists. Procurement teams should review the conditions mentioned in the contract for limits and restrictions which may be imposed due to the age of the part. If the terms allow, return the excess, undamaged, and usually unopened parts to the supplier and request a credit of the amount originally paid. To minimize any reverse logistics fees, align this process with an existing delivery or service schedule with the supplier.

Supplier buy-backs, where allowed, are the preferred method to reduce inventory. Not only is space recovered from the storeroom and the carrying cost of inventory reduced, but companies may recover up to 100 percent of the original cost of each item returned.

Vendor Managed Inventory

Leverage Vendor Managed Inventory (VMI) or Consignment Inventory while negotiating sourcing contracts with a supplier. Essentially, the supplier assumes the role of inventory planner for the customer and the asset is owned by the supplier until it is consumed. Instead of the customer reordering the item when supply has been exhausted, the supplier is responsible for replenishing and stocking the customer at appropriate levels.

Interfacility Transfers

If the organization has multiple facilities and/or more than one storeroom — and the data is clean enough to permit accurate assessment — consider implementing an interfacility transfer process. For businesses operating common assets across facilities, this can prove to be highly efficient and cost-effective. In this scenario, if one storeroom has an excess quantity of one or more items and another storeroom needs the part, it could make sense to reallocate the part(s) to the location in need. Limitations may apply to companies with poor master data management practices or those that have not instituted a universal item number across locations.

Asset Resale or Disposal

Eligibility for either resale of parts or disposal may be limited by region, country, or state/province. Business partners may also have restrictions on the items they’re willing to receive based on whether these items can be resold or properly disposed.

In one proven scenario, an asset reseller will offer to accept the addressable excess on a consignment basis. That is, they will inventory the items in their warehouse at their own cost. The benefits to this scenario are numerous — the items removed from the storeroom will be taken off the books of the business (since they’re no longer stocked inventory), and will no longer incur a carrying cost. In addition, valuable space will be freed up in the storeroom. The downside is that, for most companies, it will take longer for the business to receive funds back from the sale of the assets, as funds are paid following the successful resale of the items.

One recent engagement saw a return of approximately 25-30 percent of actual cost after LTL shipping fees and partner’s commission were counted. This compares to single-digit returns from bulk asset resellers and is impacted by the reseller’s fee, shipping costs, and item condition.

Unlike consignment programs, bulk asset resellers offer to buy the items as-is for a lower price. The benefit to this approach is that companies can often extract more items from their storerooms.The downside is that the commissions, while paid nearly immediately, are far lower compared with consignment resale partners.

Alternatively, companies can trash or recycle excess inventory. However, this is not a preferable option, not only due to waste management costs and a zero-percent commission but also due to the downstream impact on the environment. This approach should be adopted only if there are no other options or the items in question are unsellable due to damage.

The physical removal process requires planning and coordination between the cross-functional team and the storeroom resource(s), as well as third parties such as the asset resellers or suppliers accepting buy-backs. The team will need to estimate the need for and acquire materials such as gaylord totes, intermediary boxes, packing tape, and pallets coupled with a copy of the list of parts flagged for removal.


Experts recommend monitoring at least two parameters — one focusing on working capital and the other on service quality — to track performance and effectiveness. These include:

  1. Inventory turn
  2. Downtime due to spares unavailability

Inventory turn is defined as the ratio of annual consumption value to total spares inventory. The inventory turn ratio of top-quartile companies is in the range of 1.8–2.0, while the medium-quartile companies are marginally more than 1.0, and the lower-quartile companies’ ratio hovers around 0.5.

Production downtime is largely monitored and tracked by the manufacturing team — any downtime attributable to unavailability of spares affects credibility of the strategic inventory management team. This team should perform root cause analysis or Why-Why analysis to understand the reasons and adjust the reliability engineering algorithm to factor them in.

There are other allied parameters such as new spares addition, number of expedited orders, value of slow-moving and non-moving parts, etc. that could be tracked to get a better handle on the inventory. However, it is recommended to start with the basic parameters and consider monitoring other parameters depending on availability of the resources.


While companies can achieve 30–40 percent reduction in inventory, without robust policies and governance models the inventory will revert to its original state and the benefits may be negated. The bare minimum is to set up processes to streamline the addition of new spare parts and build the ability to challenge and work collaboratively with project teams comprising engineering, maintenance and storeroom operations.

As a best practice, companies can develop a new spares recommendation form to be completed by the project engineers and sent for review by the strategic inventory management team. This team uses the reliability engineering philosophy and challenges the suggested safety stock (min/max) whenever needed. Safety stock for these new spare parts can then be set after achieving consensus with the cross-functional team.

The other key governance process is to periodically optimize the existing inventory based on consumption history and to consider the recently created items that were not part of the original analysis due to cut-off date. Based on the results, the safety stock levels (min/max) should be adjusted and activities such as inter-plant transfers, physical removal, asset resale or disposal, and supplier buy-backs should be executed.

A Concerted Effort to Manage Change

Change is an ongoing process. Managing this change requires a rigorous effort with multiple groups. The key member groups that are affected by this change are engineering, capital project owners, plant maintenance, storeroom operations, strategic inventory management, reliability engineers and procurement. To effectively embrace this change across the enterprise, it is essential to reframe the mindset by addressing individuals’ behavior, attitudes, and emotions. As it has been proven, a system can change only if there is an external event/trigger(s) — training and group discussions are often found to be effective triggers. The training sessions should focus on:

  • Introducing the algorithm described in this paper, explaining what it encompasses, the operating philosophy and principles
  • Discussing the detailed methodology, benefits, checks and balances, and timelines
  • Conducting participative discussions on assumptions, risks, the principles underlying the algorithm, etc. with representatives from each group — allowing them to shape a desired service level and jointly take ownership of the processes and methodology


Adoption of a fact-based spare parts inventory management philosophy — not one fueled by “gut feelings” and unsubstantiated assumptions — will lead to a more appropriate stock level outcome, reducing both current and future excess. Based on reliability engineering, these strategies and tools yield positive, quantifiable results for organizations with spare parts inventory on hand. More than mere short-term cutbacks, enabling a conscious culture and set of policies and procedures inclusive of cross-functional teams and sustainable governance will drive ongoing, appropriate inventory levels.

The process may contain certain challenges, not the least of which will be obtaining relevant and usable data. “Clean” data is desirable but not mandatory for the project, as it will not only allow for clear visibility into excess parts but also deliver the right-sizing of stock levels. The cross- functional team will be required to review and interpret the data inputs and outputs, converting them into actionable plans for the organization. Careful planning with all stakeholders — from data analysis to physical validation and removal — drives meaningful results, offering increased capital, cost reductions, and freeing up of valuable storeroom space. Moreover, interplant transfers, supplier buybacks, and shipments to third-party asset resellers may reallocate parts to other locations and/or return additional capital to the organization.

Exact measures of success and their association with specific project milestones should be defined by the core team at the onset of the project. Irrespective of how they are defined, there are tangible and realistic benefits for companies looking to optimize their spare parts inventory. A thoughtful mix of information, people, policies, and procedures will ensure that these goals are met appropriately.