Monetizing Data in Maintenance: Data-driven Spare Parts Management
Management of spare parts and other materials needed for realization of maintenance processes is one of the key functions in physical asset management. Especially in power generation, oil and gas and heavy chemical industries, spare parts inventories can easily add up to tens of thousands of various items, in a value of hundreds of millions of euros.
Efficient spare parts inventory management can have significant impact on the financial performance of the company. Better spare parts management can lead to improvement of financial performance of the company. Spare parts inventory can lock in significant amounts of working capital. This article summarizes recommendations for effective spare parts inventory management and spare parts optimization using various sets of data and statistical analytical methods.
1 Eight rules of good spare parts management
In our previous research, we refined the following eight rules – best practices – for good spare parts management:
- ) Focus on preventive maintenance – for preventive maintenance no inventories of spare parts need to be held.
- ) Solve problems in spare parts processes.
- ) Segment your spare parts portfolio.
- ) Evaluate spare parts criticality.
- ) Apply suitable forecasting methods and verify their accuracy and reliability.
- ) Use special methods for intermittent demand items.
- ) Treat your master data well: Identification and naming of spare parts
- ) Consider the whole lifecycle of your assets while making decisions related to spare parts.
In this issue of Maintworld we will focus on forecasting – the essential element in inventory management. Follow the rules 5 and 6 to apply suitable forecasting methods and use special forecasting methods for spare parts with intermittent consumption. Add rule 7 to improve identification and naming of your spares in master data.
Spare parts management starts with good forecasting
The next step in the specification of optimum spare parts inventory management regime is the prediction of future consumption of the items in stock. The forecast is always based on transactional data from information systems – history of spare parts consumptions, which must be representative (meaning sufficiently long). In the case of spare parts, we usually work with a history of three to ten years (depending on industry). Three years of recorded history seems to be the minimum for intermittent items. A general rule here applies: the longer the history, the better and more reliable the forecast.
When analyzing historical consumption, we need to carefully distinguish between material consumed for planned maintenance (planned shutdowns, turnarounds, preventive maintenance) and spare parts issued for unplanned (corrective) maintenance – repairs. In forecasting, we must adjust the history for planned maintenance.
In the forecasting process, items should be treated individually, according to the character of their consumption. Items with common demand patterns (high runners – fast moving items like fasteners, etc.) can be forecast using a number of standard statistical methods normally used in inventory management (moving average, exponential smoothing, Holt’s exponential smoothing, trends, seasonal indexes, Winter’s method, etc.). Items with intermittent demand require a special suitable method to be applied. The use of standard methods of prediction and inventory management in case of intermittent items results often in a substantial overestimate of future consumption and therefore excessive inventory level.
Intermittent demand is the pitfall of spare parts management
One of the specific problems in spare parts inventory management is the nature of spare parts consumption - intermittent demand. If we analyze the consumption history of a typical spare part, we find that the historical consumption in most of the analyzed periods amounted to zero. Such infrequent or intermittent demand, usually with demanded quantity of just a few pieces, is very typical for spare parts and other maintenance inventories. An example of the consumption history of an intermittent demand item is presented in Fig. 9. In maintenance, intermittent demand is quite often combined with long supplier leadtimes. For maintenance inventory management, intermittent demand and long leadtimes are a tricky complication, often leading to large overstock. The main problem with managing and forecasting intermittent items is that the standard forecasting methods used for fast moving goods (for instance moving averages, exponential smoothing, Holt’s and Winter’s method, constant or regression models with seasonal indexes, etc.) simply do not seem to work for these items. In case of intermittent consumption, special statistical methods (such as bootstrapping or Smart-Willemain method) need to be applied.
Smart and Willemain (2004) suggested a stochastic simulation forecasting method. Using this method it is possible to specify minimum inventory level (re-order level) in order to ensure fulfilment of requirements with target probability (logistic service level, target of availability). The method is based on random sampling from the history of consumption. In statistics, similar methods are called bootstrapping.
Besides intermittent items, in a large maintenance inventory we can also find fast-moving items with stable and high regular consumption. These are especially items of common consumables like fasteners, generic gaskets, or bearings. For these items, standard methods of inventory management and future demand forecasting can be applied.
Identification of spare parts and cleaning master data
In spare parts optimization projects, we quite often face various problems with quality of spare parts master data – especially naming: issues with unstandardized naming or incorrect names, names in various languages, different word order, typos etc. hinder significantly all efforts in spare parts optimization and generally result in duplicate (or multiplicate) master data records for identical materials (identical spare part is stored in several master data records with different names). In order to clean spare parts master data, we apply advanced data analytics on master data to identify or cluster duplicate (or similar) items. This key analytical method is “matching” – comparing names and certain master data attributes to assess similarity of spare parts. For this we use multicriterial comparisons. To analyze spare parts’ names, metrics like Levenshtein or Hamming distance are used and combined with triplets analysis (comparing all triplets-threes of characters found in all names in master data).
The result are clusters of similar or duplicate items found in master data. These groups of highly similar items are given to maintenance engineers for validation. After duplicate items are confirmed, master data can be rectified – correct items is selected to be used and all duplicate items are erased or deactivated. Examples of Matching are presented in Figure 8 and Figure 9.
Good spare parts management has significant impact and benefits
It can be concluded that spare parts management as a part of physical asset management has significant impact on financial statement of the company. Good spare parts management brings the following benefits:
• Optimum spare parts quantities are purchased
• Optimal purchasing cashflow
• Lower inventories
• Less unused inventories
• Higher availability of needed spare parts
• Good risk management
Various sets of data can be used to support or optimize spare parts management, especially in spare parts segmentation, criticality assessment, forecasting, master data rectification. The examples demonstrated in the paper indicate ways to utilize vast amounts of data available in organizations today – to monetize the data by improving efficiency and effectiveness of spare parts management processes.
The occurrence of unwanted emissions is a headache for the process industry because of the loss of raw materials, safety risks and environmental damage.
Reliable condition monitoring secures transition to carbon-neutral energy at Oulun Energia's biopower plant
Air quality is in sharp focus for a range of industries. Monitoring air pollutant emissions is essential for many reasons.