Identifying and analysing the
critical success factors for
manufacturing efficiency —
for sustainable, on-time and
in-full results.

22^nd February 2015

Introduction

Despite having been mined extensively for efficiency savings over previous decades, the manufacturing sector continues to face the same old challenge. Since competitors in a mature landscape use price as a tool for winning market share, and as customers demand lower prices on products, manufacturing businesses must continue to identify and extract new efficiencies from their teams, production systems and equipment.

Even the most capable owners, managers, consultants and production teams, in searching for and fine-tuning methods for increased productivity, continue to encounter one fundamental challenge: difficulty in accessing the data they need quickly enough, and in an appropriate format.

The success of production teams in streamlining manufacturing, and in delivering on-time and in-full fulfilment results, appears to depend on three key factors; what has happened, what is happening, and what needs to happen. This white paper explores these factors and investigates methods for successfully capturing and delivering the right information quickly and efficiently.

The three ‘H’s of productivity heaven

While 100% efficiency may be overly ambitious for a production manager, 80-90% is achievable for long periods of time. Any process which achieves an efficiency of close to 90% is likely to be performing better than their competitors.

If a production manager knows what has happened, what is happening and what needs to happen, they can see 90% as a realistic goal. They can make well-informed decisions, can correct problems as they occur, and can put new practices in place to ensure the problems don’t reoccur.

1  What needs to happen?

This is where the decision process is most critical and pivotal to success.  It may seem like a simple question, but this is where most errors occur – and where getting it wrong can be most costly. Unfortunately, most managers spend their time managing current issues rather than planning on avoiding the next sequence. 

What needs to happen in the majority of manufacturing situations is as follows:

• Source the materials

• Identify the processes route

• Schedule the activities

• Ensure that resources are available to fulfil those activities. 

This seems simple enough on the surface, but add in other factors (batch sizes, common parts, lead-times, multi-level bill of materials and third-party sub-contractor variables) and suddenly it becomes much more complex.

Most production teams would agree that their role is to ensure that activities are carried out in a timely manner, and to correct the resulting problems when they aren’t. However, many teams don’t have access to data which can tell them when there are problems, so they can’t circumvent them.

In the early days of technology, big companies would purchase expensive systems to help them plan production, while SMEs would have to work from gut instinct and experience. Nowadays, however, affordable technology is available to help any size of business in planning their production. At this point, it may be useful to consider what is available and what businesses should be doing, as a minimum, to deliver efficient planning and to provide production teams with what they need in order to make decisions.

Material requirement planning (MRP)

MRP systems are hardly new, but not all of them can simultaneously meet three key objectives:

• Ensure that all materials are available for production

• Maintain the lowest stocks of materials and products

• Plan the manufacturing activities to be carried out.

In its simplest form, the MRP produces two outputs: a production schedule and a purchasing schedule. It must factor in bills of materials, lead-times, batch sizes, common parts and due dates. 

One of the biggest problems with MRP systems is the tendency to make planners lazy. It becomes too easy for planners to accept whatever the system suggests, despite the many potential pitfalls. If poor data is entered into the system, its outputs cannot be reliable. Input errors occur in inventory levels and cycle times, with defects and demand plans affected. It would appear that the biggest impact on MRP output errors is the keying-in of incorrect data, causing miscalculations (some of which can be catastrophic). These errors can be minimised (see below), but any good MRP system should not only suggest at the planning stage what needs to be done but should also demonstrate how it derived its answers.

Usually, planners will have a gut feeling for what is happening in the factory and errors can normally be spotted thanks to a combination of human vigilance and supporting data (for example, when a customer orders 10 times more than usual).

The other major drawback of MRP is that it often takes no account of capacity. As a result, it can deliver results which are impossible to implement due to manpower, machine or supplier capacity constraints. Production scheduling tools can be very helpful when dealing with this drawback.

Shop-floor loading/scheduling/capacity planning

This concerns balancing demand with the resources available. Whist it might be ideal to have an infinite amount of capacity to allow the MRP to be the final stage, this is rarely the case (and wouldn’t make for an efficient factory). 

A shop-floor loading tool, sometime called capacity-planning software, provides a reliable method for allocating resources to activities. The complexity of the shop-floor planner should depend on what the business feels is an important resource to plan. For some businesses, this will mean focusing on the most expensive or restrictive resources (such as a piece of machinery) as the only requirement for the plan, while others may require materials and people to be factored in.

Like MRP, good planning software should be suggestion-driven, rather than forcing a plan: human intelligence should not come second to the computer. A good planner will know what is possible and, under certain circumstance, which is the only way to achieve a job – even if the parameters suggest that this isn’t the best or most efficient practice. 

Too many finite-capacity-planning tools are too rigid. In one example, a job was configured to run on a 100T press (the target machine group). The planner was unable to move the job onto a 300T press because the 300T was not profiled for the job, and the company didn’t want the system changed to what was deemed to be an inefficient system. However, when the 100T press broke down, this became a viable option. In this scenario, the entire planning system fell apart, as it could not reflect what was achievable in the real world.

The best planning tools give the planner as much information as possible from which to make decisions. One very effective system allows the planner to load jobs as they see fit; the system offers the planner all the information they need, shows capacity and stock availability, highlights late-running jobs, and manages priorities and predecessor job statuses and importance. The planner simply drags and drops items onto a grid – a traffic-light system informs the planner of any conflicts/problems or better options, and can deliver a work-to list for each day, week or month.

Of course, all the best plans can be subject to change, and rescheduling is commonplace. A robust planning tool should be flexible enough to change the plan and quickly visualise the impacts of those changes; this becomes increasingly important as the run-date draws closer. When planners have this vision, they are able to make the best possible decisions.

2  What is happening?

A good production manager will have their jobs scheduled and their materials on order; now, they need to keep an eye on what’s happening. This involves visibility of the processes and ensuring that they adhere to the plan. If they deviate, it’s essential to know what impact this has and to have options as to what actions to take. Under normal circumstances, the more indicators the better. However, on busy shop-floors this can lead to information overload. This is where colours and visuals become important tools in raising issues and making them clearly understandable. 

The first stage of a process usually begins with the arrival of materials, ensuring that they are identifiable and that they are placed in a location where they can be used to satisfy production.

Chaser reporting

Stores teams are usually hands-on people; some can use technology, but they tend to be in roles which react to requests dictating what should be done. The simplest and most effective way to ensure efficiency here appears to be giving the stores manager a daily/hourly or weekly chaser report. 

Depending on budgets, some companies like their chaser reports delivered electronically to handsets. However, a printed or emailed list of materials due to arrive, materials required for production and earliest/latest timescales can be a very effective way to increase the stores teams’ efficiency. 

Providing this information to the stores enables them to be proactive in ensuring that materials are where they need to be. It is not uncommon for, say, a factory to have four lorries arriving at exactly the same time, with panic ensuing.  Some companies can manage this with the suppliers; if that isn’t possible, knowing in advance what is going to happen can help to avoid that panic.

Another major problem affecting flow is the supplier’s ability to deliver on time. Even the best plans are dependent on the timely arrival of materials. The answer is to treat suppliers as if they were employees, and to send them a chaser report specific to them: “This is what you should be supplying, and when.” A good system will be able to send them regular, automatic notifications of expectations, removing any excuses on their part.

Shop-floor data capture

Astonishingly, some businesses still issue hand-held tickets for jobs. The operator fills them in, and then they are keyed back into the system at a later stage. This creates substantial unnecessary processing, a wide margin of human error, and significant delays in access to critical information. For a relatively low price, this data could be captured electronically at source, from the shop floor, and fed back into a system which delivers invaluable visual indicators of what is happening. Simply showing the status of running jobs in a factory, and indicating which resources are being used, can enable a business to save up to 20% in utilisation, to reduce errors, and to resolve issues more quickly. 

All too often, there is a disconnect between the workforce and management. If management have visibility of a problem, they are in a better position to resolve it. The production team may only be interested in resolving that particular job; if a manager sees a problem, however, he or she has an interest in the bigger picture, and can see the impact on the plan. It may be that job was a ‘filler’; rather than delaying important jobs, the manager can decide to continue a process elsewhere to prevent a bigger problem. 

A colour coded system, displayed on a graphical representation of the shop floor, is a simple but hugely effective solution. It provides an instant display of the current shop-floor status which is easy for anyone to understand at a glance.

One of the simplest ways to collect data, which is also fast and very cost-effective, is with fixed terminals located around the factory. Despite concerns about operators having to queue, this doesn’t happen and operators are able to sign on to jobs, record down-times and reasons, and record production quantities at the end of runs with little impact on their normal duties. This also captures useful information about machine usage, highlights common problems, and calculates operator performances. It also helps operators by providing access to important information such as quality concerns, pictures of finished parts and stock locations – all helping to reduce problems.

Machine-monitoring

For long-running batches, it may be worth implementing a machine-monitoring solution. This can be a little more expensive to set up, but connecting feeds to machines not only reduces keying errors but enables corrective action on items such as slow or fast run-rates. These alerts usually highlight quality issues, or inaccurate assumptions, very quickly.

3  What has happened?

Many methodologies promise continuous improvement, but the important issue is knowing what has happened. Knowledge is power when it comes to making improvements to the process, so it’s vital to learn from past lessons. Recording as much critical information about the process run, as simply as possible, is crucial. Equally important is the ability to locate, access and process that data quickly. 

Scheduled reports, dashboards and exception-reporting are excellent methods for getting to the data quickly and easily, and one of the most effective ways to increase productivity is by publicising this performance data. This helps to incentivise teams, enabling them – and suppliers – to assess their own performance levels. Displaying this information on dashboards around the factory is a simple and cost-effective way to incentivise teams to come up with new ideas and to tweak their processes.

Actual costing and traceability are sometimes forgotten when measuring performance, but they are critical to profitability. At the start of the process, an estimated cost is surely derived – but how accurately depends on the procedures and cost importance. However, measuring this against actual cost is more important than the method, and knowing the impact of production problems helps to justify any costs required to implement a process change.

Tagging information at batch level (whether for material or production runs) is an ideal way to identify where problems have occurred. It may take several months for a shipped product to fail, so the ability to review the process, and identify the cause at batch level, will allow you to identify the number of affected parts. Minimising recalls can save a business millions of pounds in costs and penalties.