By Philip Hanekom, Associate Consultant at Pragma Africa
Any Asset Management (AM) system either fails or succeeds because of people. Competence, process discipline and capacity are key, and without these fundamentals, in place, any system is bound to be ineffective.
This article will focus on capacity, but first, let’s take a quick look at the other two fundamentals. A discussion on competence deserves a whole article to itself, but in short, it is about ensuring employees have the ability, knowledge and motivation to make decisions in their position. A lack of process discipline results in a reliance on individuals as knowledge holders, fragmented communication, a breakdown in systems, a lack of clearly defined roles and responsibilities, and frustration on all fronts. It’s no mystery why a lack of process discipline leads to ineffective systems.
Back to capacity. For any initiative, capacity in an organisation is imperative. As with competence and process discipline, it is no easy task to define resource capacity requirements optimally. As a result, many organisations end up either over or under capacitated. With no clear understanding of how to correct the situation or evidence that capacity itself is the real underlying issue, things are unlikely to change, and the AM function suffers as a result. It’s, therefore, necessary to develop a model with which capacity requirements can be determined.
Before delving into numbers and calculations, it’s important to get the fundamentals in place that will inform the model and any decisions based on it. At the very least, an AM policy should be drafted that defines the organisation’s expectations of the AM function and sets up the AM function in a manner that supports the organisation’s objectives.
Some of these policy objectives may influence the maintenance capacity model by stating key requirements in terms of outsourcing and operator asset care. These factors will obviously have to be considered in any model.
A maintenance function could be capacitated by the owner, contractors or a hybrid model. In addition, some of the maintenance responsibility could be transferred to the asset operator. This factor alone indicates the critical need for a formal policy document.
When it comes to understanding optimal resource requirements for a maintenance organisation, there are no shortcuts. Assets have inherent reliability, and the associated maintenance requirements to ensure that level of reliability have largely been determined by the design of the asset. Short of changing the design, there is very little that can be done to significantly alter the maintenance requirements of an asset. For this reason, before even considering the number of tradespeople to appoint in a plant, the maintenance requirements of the whole system need to be well understood. It is mostly this – the desired reliability (within the limits of the design) and maintenance required to achieve this – that will determine the number of tradespeople required.
Let’s look in detail at the following key information required to make an informed estimation of trade capacity:
- Plant operating model
- Resource planning, considering planned staff utilization and effectiveness of maintenance work planning over the short and longer term
- Tactical maintenance workload
- Average non-tactical maintenance workload
Plant Operating Model
The operating model largely determines the intensity of maintenance required. A plant operating 24 hours a day/7 days a week/365 days of the year sweats its assets more so than a shift a day/5 days a week operation. Considering an asset requiring a maintenance intervention every 500 hours, for a 24/7 plant, maintenance would be required every 3 weeks. For a plant running a shift a day for 5 days a week the intervention would only be required after 63 shifts, or after just over 12 weeks of production. This simple example illustrates how the same plant, running at different planned utilisation levels would require different maintenance resource capacities, in this case 75% less capacity for the less utilised plant.
How maintenance resources are used is another big determinant of the capacity required. Two things to consider are how resources are planned to be utilised and how effective maintenance work is planned.
In order to determine the number of hours a tradesperson would have available per year to perform maintenance work, it’s important to consider how that person is planned to be utilised. This is typically a strategic decision and would consider basic working hours, the personal development policy, meeting structures, leave policies and other allowances. As an example, let’s consider a typical working year. Of 365 days a year, 104 are weekend days. This leaves 261 working days on which a tradesperson could work a shift. Factoring in leave of minimum 15 days per year, 12 public holidays and an estimated 5 days’ allowance for training, illness, family responsibilities, etc leaves 229 working days, i.e. shifts, per tradesperson to do maintenance. Given the shift length it’s possible to use this information to determine the hours available per year that would cover the maintenance requirements of the assets.
One final thing to keep in mind in resource planning is that time is lost during a shift as well. Planning for this is essential, and it includes things like lunch and other breaks. Unfortunately, the more difficult factor to plan for is how effectively maintenance work is planned. This impacts wrench time, which is the time a tradesperson spends effectively performing maintenance, excluding time wasted on travel, sourcing spare parts, etc. A fair estimate of wrench time for most industries is about 35-45%, based on over 178 assessments conducted by Pragma. The wrench time applied to the annual tradesperson capacity would give a good indication of the amount of maintenance work that person can realistically complete per year.
Understanding the capacity of a tradesperson is useful. But if, from a planning perspective, the maintenance workload is not effectively balanced over time the team might be able to, on average, complete all maintenance work for a given period, e.g. one year. However, the team will be short-staffed during the peak maintenance demand periods and overstaffed during the troughs. This means that during peaks, maintenance work will fall behind and potentially be completed at a time when the maintenance demand is lower, which could very well be too late, considering the failure patterns of the asset for which maintenance has been delayed.
If the load does spike at certain times and this cannot be avoided, consider staffing for the norm and contracting in extra capacity during periods of high maintenance demand.
Very simply put, maintenance work is either expected or unexpected. Expected maintenance, if it has been developed from a thorough understanding of component failures, would be determined by the appropriate tactic selected to prevent or mitigate the impact of that failure. It’s therefore referred to as tactical maintenance.
Tactical maintenance should be routine and well understood, even if it is based on OEM recommendations or experience. The task specifics, tool and spare requirements, trade allocation, cost and duration of the work should be defined and configured in the Computerised Maintenance Management System (CMMS). The duration and trade allocation can give a very good indication of the typical maintenance workload and is therefore essential for determining the plant maintenance resources required.
Unfortunately, tactical maintenance isn’t the only demand on the maintenance team. Unexpected maintenance is a reality and while it’s always good to aim for zero failures, it is prudent to still plan for asset failure. Here, history is the simplest guideline. The average availability over some period in the past may be used as a predictor of availability for the future. Alternatively, where a new plant is being developed, Reliability Block Diagrams (RBD) can be used to predict the system availability and determine critical components for system reliability. This, however, is significantly more complex and time-consuming.
Given the maintenance demand over a period, e.g. a year, it becomes simple to calculate the resource requirements based on the expected capacity of a single tradesperson. If the demand per trade is known the number of tradespeople per trade can be calculated as well.
Asset Replacement Value
A simple way of ensuring the calculations deliver sensible results is to compare the overall planned or actual maintenance cost to asset replacement value. Maintenance cost as a percentage of asset replacement cost is a good indicator of how effectively maintenance and asset lifecycles are being managed. Excessive maintenance cost relative to asset replacement value could indicate either ineffective maintenance or ineffective asset replacement strategies. It is important to note that the ideal range is dependent on the equipment age, maintenance and other management practices, and the asset intensity of the organisation. Pragma’s AMIP Framework is a five-stage maturity framework. It defines the maturity of organisations’ life cycle management practices in terms of maintenance cost as a percentage of asset replacement value, where maturity benchmarks for various industries in varying degrees of maturity have been defined.
A large proportion of the maintenance budget is often spent on labour, so the cost of over-staffing can be significant, but under-staffing the maintenance function and not being able to perform effective maintenance can lead to an increased risk of costly downtime. So, it’s important to consider the staffing requirements (based on the approach described in the article above) in the context of the organisation’s objectives and operations in order to balance the risk and cost with the required asset performance.
Furthermore, a key benefit of a formal resourcing study is that various decisions can be made on an informed basis, such as outsourcing or insourcing decisions, operator asset care policy, and identification of maintenance demand spikes or troughs. Common sense should prevail, but the framework for informing the staffing decisions, as described, can prove to be very useful as a decision support tool.