Diagnosing problems in your Manufacturing Planning and Control (MPC) systems and permanently fixing them

Part 1 Identifying the real problem

Symptoms of the problem could include: Excess inventories, obsolete stock, late deliveries, shortages of components or raw materials, poor first time pick rates, cash problems, poor lead-times, which for strategic reasons you need to reduce, excessive expediting or transport costs.

This first of two articles on permanently solving problems in MPC systems discusses the problem of separating the symptoms from the real problems. The second article will discuss the methods of prioritising the issues arising from this analysis and solving the problems permanently. The two articles were reproduced by the UK Institute of Logistics and Transport in their house magazine "Focus" in May and June 2000.

Links to other best practices and training at bottom of page.

Firstly when was the last time you had to conduct a crash exercise to repair the damage caused by customer complaints for poor delivery performance, or complaints from your bank manager because your working capital was unacceptable, or worst of all, both?

Or maybe you had a problem, which seemed to go away without anything fundamentally changing. However the problem mysteriously reappeared three months later. What probably happened in this case was that demand has fallen, the mix has reduced, or capacity increased to temporarily take the pressure off the planning and control system as shown in the following diagram.

Or you thought you had solved the problem. But it reappeared later. In this case you probably solved the symptom of the problem not the cause. I can recall a purchasing manager who was fired for constant component shortages. Afterwards (when the problem reappeared) the real problem was diagnosed to be that sales orders spent 2 months in the sales department (for contract review) before planning whilst the purchasing lead-time was being eaten into. (The customer delivery date had been fixed earlier based on realistic lead-times.)

Secondly what did you do when this last happened? If there is an accident you need to do two things:

• Clear up the wreckage
• Stop the accident happening again

My guess is that you made a great job of clearing up the wreckage but were too busy dealing with the next accident to prevent this one happening again. If everyone is fighting the alligators who is draining the swamp?

It is important for your accountant to recognise that they do not control working capital, your MPC system does. Your accountant can only delay payment for goods and services that the planning and control system ordered too early or for which there is no demand from your customers. When you delay payment you go to the bottom of the suppliers' priorities, which causes you to hold more stock to defend against that! (See Malpractice 001: Not Paying Suppliers On Time.) A better answer is to buy what you need when you need it and pay for it straight away and remove the accounting administration involved with delaying payment.

If you promised your customer a delivery date, it is perfectly reasonable for them to expect delivery then and the things that are not acceptable are:

1. You could not give an accurate delivery promise
2. You did not tell them that you were going to be late before you were actually late

The problem is usually not a lapse of concentration but a fundamental fault in the planning and control systems surrounding manufacturing or purchasing.

In my experience most MPC systems have fundamental faults which are relatively simple to fix if only you could separate the symptoms from the causes. You know you have symptoms of a problem but what are the causes?

Like any good GP you look at the symptoms first:

• Poor customer delivery performance / first-time picks from the warehouse?
• Too much stock, much of it obsolete?
• Too much shortage chasing?
• Too many component shortages?
• Lost business?
• High transport costs?
• Etc.

OK you recognise the symptoms. What next? Do not jump to the immediately obvious solution. This usually results in treating the symptoms not the causes.

The following diagram summarises the process (with apologies to Dr W Edwards Deming)

The Diagnosis

Root Cause Analysis

Try asking the question "why?" five times first (as in the following example):

Qu. Why have I got too much stock?

Answer: I have bought more than I have sold.

Qu. Why have I bought more than I have sold? (Do a "Pareto" analysis of the high stock holding costs)

Answer: One of the main reasons is that we forecast raw material requirements but we do not end up selling it.

Qu. Why do you do that?

Answer: We cannot forecast customer demand accurately.

Qu. Why not?

Answer: We use historical demand smoothed exponentially but it does not seem to work.

Qu. Why not?

Answer: We do not talk to the customer after we have forecast to try to validate the results.

Maybe that is the answer!

Pareto Analysis

Pareto analyses may be applied to any frequency distributions to highlight root causes for:

• stock-outs
• quality problems
• high stocks
• lead-time variability
• late deliveries
• reasons for lost business
• etc.

You may see some symptoms here in your analysis of reasons for lost business. In the longer term you may be able to reduce lead-times or improve quality etc., but the MPC system needs to react to the current mismatch now.

Next try a policy check

What are your policies relating to manufacturing, customer service and stock levels? Often current policies cannot be justified in the light of current circumstances.

Ask the following questions:

1. What is the ratio of customer required lead-time vs. process lead-time?
2. What is the ratio of customer required lead-time vs. distribution lead-time plus process lead-time plus purchased parts acquisition time?
3. What service levels (first time picks) am I trying to achieve?

These questions are asked by product, product group, customer, and market because:

• Some products may be supplied in the required time, whilst others may take longer.
• Some customers may not be so demanding for some products.
• It is a fact that not all customers are equal.
• Different product / product groups have different demand variability and customer service level requirements.
• Some markets are more demanding than others.
• Consolidation of demands across customers can have a smoothing effect on demand variability.

It is important to try to at least match the competition, or beat the competition by product if you are differentiating your products on lead-time.

If customers want things quicker than you can buy and make them you need to hold stock at some stage in the supply chain to satisfy the mismatch (starting with the longest lead-time processes and using "Lead-Time Analysis"). This policy review may result in offering some products ex-stock and others made-to-order.

4. What stock policies are in force?

Are all products / components equal? They do not need to be. In fact it is beneficial for them not to be.

• Long lead-time items need more stock.
• You can afford to hold more stock of the inexpensive items.
• It is difficult to stock bulky items.
• Shelf life may be a problem to some components.
• Variable supply problems need more safety stock.
• Ex-stock items need finished stock.

Next try a business process check

• Do we have a way of capturing and assimilating customers' demands to allow us to plan accurately and on a timely basis?
• Do we have a way of determining internal and purchased direct and indirect material requirements from the demand and is it effective?
• Do we have a way of determining internal and purchased capacity requirements from the demand such as by using Takt time or Skill matrices and is it effective?
• Do we have an effective stock, WIP, and on-order at suppliers, recording process? Is it timely, relevant, accurate and data captured where the action occurs? (See Previous Best Practice of the Week 44: TRAP)
• Do we have an effective capacity availability recording process? (Skills & / or equipment, internal & / or external)? Is it accurate and timely?
• Do we have a way of keeping WIP and finished, sub-assembly, and raw material stock to a minimum and is it effective?
• Do we have an effective way of communicating demands to suppliers? Is it accurate and timely?
• Do we have an effective way of handling product life cycles and product specification change?
• Do suppliers deliver good quality within their quoted lead-time?
• Is the planning system itself effective? Does it produce do-able plans and accurate materials requirements from the demand, which handle variability in demand?
• Is there a named owner for the processes and the information?
• Maybe there is a process missing or ineffective, or it is not owned by anyone? Are roles and responsibilities adequately defined?

Notes on control systems

1. Planning and control have an effective horizon. Long-term strategic and short-term tactical. Examples would be:

• Short-term: Do we need overtime to get that job out on time?
• Long-term: Do we need new capital equipment to cope with the sales forecast?

2. Do we act on the information? For each process is there a valid control system in operation? I.e. are the following elements of control visible?

• A Plan
• A Tolerance of acceptable deviation from plan before control action will be taken
• Measurement of actual performance
• Feedback of deviations from plan fed back to the person with the responsibility and authority to act on it
• Actions arising from the feedback

Next try a planning parameter check

• Are the demands right and how do you know?
• Are the list of parts / ingredients (Bills of Material) you are using for planning the same as are being used to make the product and how do you know?
• Are the lead-times for production and purchasing right and how do you know? Are there any step functions in lead-times, and what are the trigger levels for these? When did you last ask the supplier?
• Are the safety stocks right and how do you know?
• Are the re-order levels right and how do you know?
• Are the scrap / shrinkage allowances right and how do you know?
• Are the batch sizes the least you can achieve and how do you know? (Often actual batch sizes are smaller than planned batch sizes anyway because shortages cause batch splits.)
• Are your records for stock, WIP, and on order at suppliers, right and how do you know?
• What is your capacity and that of your regular suppliers and how do you know?

Do you maintain the above information sufficiently accurately and on a timely basis to allow you to plan and how do you know?

How do you know if you know?

• Calculations will be based on relevant mathematical principles and current information.
• You will have a data collection and feedback system in operation with performance indicators to tell you. This system will measure the current performance and plot trends.
• You will be regularly auditing the procedures and calculations, and especially the policies.

Culture

Occasionally an atmosphere exists which is simply not conducive to beneficial change and may present a significant obstacle. I have found that in order to overcome these problems I have to go up a level in the organisation to determine the cause, which can be rooted in personnel policies and practices, lack of vision, lack of goal setting, communications, or simply a poor manager. (See Culture Development Methods.)

The Outcome

The result of the diagnostic exercise, which may take several days, may be up to 100 issues, problems, or opportunities for improvement, which all require attention! Do not be tempted to dive in and start fixing them yet. The next article will discuss methods of prioritisation, implementation, making the change stick and making sure you do not have to do this again.

If you have fewer than 50 issues you are already world class or you have not done the diagnosis properly. (And I have not encountered too many world class performances on our initial analysis!)

__________________________________________________________

The following further best practice articles were also mentioned in this paper:

The following public training courses and in-house workshops provides solutions to diagnosing manufacturing planning and control problems:

SSC04 Production Planning & Control Back To Basics (Includes our unique 180 point self diagnostic health check which you should be able to complete following the course to determine your development needs.)

You may also be interested in the following related courses:

To discuss your consulting or training needs with one of our independent consultants or trainers please Contact Us.

 Home Page

© SM Thacker & Associates  (Consultancy and Training Specialists) April 2000 v2 August 2007