Working in a supply chain job, role or project you can inherit a whole heap of stock management problems to sort out.   If you have limited time, resources or strict deadline pressures, yet know you need to resolve and deliver solutions, which ones do you look at first and which will give you the greatest return once solved?

Pareto analysis is a principle that helps identify problems and prioritise solutions.  Also known as the 80/20 rule, the idea is that there is an 80/20 cause/effect ratio.  For example, Pareto is the idea that by doing 20% of the work you can generate 80% of the benefit of doing the whole job.  Or in the case of stock management, 80% of the problems are caused by 20% of the inventory.  The Pareto idea applied in the real world is that a lack of symmetry exists between in this case, problems and cause.  Real world figures show this although the actual percentages may vary slightly.

Applying Pareto

To apply Pareto in stock management you need to follow a few steps.

1. Diagnose and list your issues

Determine in what areas you want to solve problems in e.g. geographical areas, types of problem etc.  Scope this area carefully.

Collect information from around the business that will help you to list problems.  From meeting minutes and running workshops to reviewing business data, quality feedback forms, change requests, previous project notes and lessons learned, business proposals and helpdesk logs.  You may find your problems include:

• Inaccurate stock records
• Delivery errors
• Out of stocks or low buffer stocks
• Picking and delivery inaccuracies or
• Inventory price volatility

Note that when you collect data it must show the norm.  For example if you have stock location issues or inaccurate stock count records after a major manufacturing or warehouse system implementation you will skew your data.  If you are trying to understand lessons from an implementation project however, you need to collect data during the project, not before and after it.

2. Root cause each and categorise

Each problem must be root caused.  Techniques such as Brainstorming, Problem Restatements, Pros-Cons-Fixes Technique, 5 Whys, Cause and Effects diagrams, Matrix Analysis, Root Cause Analysis etc will help you.

After root causing each problem you may determine for example:

• Inaccurate goods receipt recording due to lack of training
• Too many daily stock adjustments due to poor timing of system updates
• Wrong warehouse stock movements due to poor signage
• Inaccurate picking and packing due to lack of training and lack of staff
• No processes for managing price volatility when purchasing stock, etc.

When completing this exercise consider if there are groups of causes, to help you categorise each root cause.  Note that when you do this however you need to group at consistent level e.g. all at the lowest sub-level or all at an aggregate level.  You can produce very skewed results if you compare for example an aggregate group such as “lack of training” with a low level sub group such as “lack of staff at 1500H in the fresh food section to do cycle counting.”  Not only will this reveal significant key contributors and potentially many lesser ones, it can also lead you to draw the wrong conclusions e.g. is it a lack of staff being trained, does the training delivery need improvement, are the training procedures inadequate, does the training time or environment need improving, are staff failing to attend training sessions due to work pressure commitments etc.

3. Score

For each problem-cause combination assign a score.  This can be based on values you choose linked with the problems you are trying to solve.  For example values can be based on:

• Number of reported errors or complaints
• Frequency %
• Stock value
• Volume throughput
• Stock volatility
• Stock budget variance

So for example if one of your “problem-cause” combinations is that you do not have the time to do regular inventory checks, you could prioritise your inventory checking time by stock value.  If you’re looking at price volatility you can consider stock budget variance by value, etc.

4. Group by root cause and sum the scores

Add up the scores for each root cause group.  Add this to a table then arrange the rows in the decreasing order of importance of the causes i.e. the most important cause first.

This will give you a priority list and by using Pareto will show you which problems to solve first, based on an 80/20 ruling.

For example if your highest priority was low quality inventory checking due to lack of staff, then you would know that solving this issue would give you the biggest benefits.  Applying Pareto theory, solving 20% of the problems will give you 80% of the benefits.

Note that you can also plot your results on a chart at this point by:

1. Putting your results in a table and arrange the rows in decreasing order of importance
2. Adding a cumulative percentage column to the table
3. Draw a graph axis putting the root cause on the X axis and percentage frequency on the Y axis
4. Plot a lined curve
5. Draw a vertical line at the 80% point.  This will show every cause to the left of the line as the most important (20% of the causes giving 80% of the benefit) and the least important causes to the right of the line
6. Add a second Y axis for the sum scores
7. Add a bar chart for each root cause using the original X axis position and the second Y axis.  This will show you the root cause number of errors in graphical format where the relative importance is strongest on the left and decreasing as you move right

If you are getting a flat curve after this exercise consider:

• Is your data correct?
• Do you have all the relevant data or are you missing some?
• Are you measuring the right problems and causes?
• Do you have the most suitable measurements?
• Are your categories appropriately divided?
• Is the scope relevant to the exercise?
• Revisit the above steps (1)-(4) before moving to step (5) below

Now is the time for the last part of the process..

5. Find a solution and implement

Starting with your highest priority problems, you need to generate ideas for solutions and may need to use cost benefit analysis, net present value or internal rate of return calculations to decide on the most appropriate course of action, ensuring they are worthwhile carrying out.

Using one of our examples you may decide that to resolve “stock accuracy problems due to infrequent inventory checks,” you cannot afford to increase staff numbers or improve technology immediately to make inventory checks more frequent.  So instead, to start with will apply an 80/20 rule to counting stock i.e. only routinely track, manage and count the 20% of stock that equals 80% of value.  You might decide that to save time and increase accuracy that for metric and KPI purposes, the remaining balance of stock can initially be summarised into a nominal cost, using nominal values to calculate cost of goods sold, gross profit, stock adjustment values etc.

These techniques can be applied in the real world showing Pareto analysis can help resolve stock management issues.

Commenting on the effectiveness of Pareto analysis in this context, Andy Hudson from Profit Through Pareto thinks “it is useful for supply chain managers to apply Pareto for stock management because if 20% of inventory represents 80% of inventory costs, a firm can go a long way to controlling its inventory costs by only tracking and managing the top 20 percent. It all makes a lot of sense but of course the 20% being tracked has to be the correct 20 percent.”  Andy continues “you can go a stage further by classifying inventory using ABC, where A lines are the top 80% of cost and 20% of items, B lines are split 15% and 30% and C lines are 5% and 50%.  You can then count and track for example based on classification e.g. A items throughout a day up to monthly, B lines monthly to quarterly and C lines quarterly to annually for instance.”

As a final piece of advice Andy states “Pareto gives a very good way of how prioritisation of effort can be guided.  I agree that in a lot of situations people do move to prioritise important things. The issue may be that each individual sees slightly different priorities – the business overall needs to set these.  One of the skills here is defining, understanding and applying these priorities.”

Tags: Manufacturing blog