Quality is not enough.

If 'world-class' quality will be assumed by demanding customers in the '90s, what else will it take to compete? The answer is a leap from part quality to Total Quality Management via plantwide information sharing for maximum economic efficiency. CIM is how to make it happen.

One of the major trends affecting American business in the Nineties is that quality, price, and on-time, even just-in-time, delivery will be expected from any world-class supplier. What started as marketing buzzwords used to differentiate topnotch manufacturers will be prerequisites in the buyer/supplier relationship in this decade.

PLASTICS TECHNOLOGY's recent survey of "world-class" molders clearly shows how far these elite manufacturers have come (see PT, April '91, p. 96). What is perhaps most startling is how many "world-class" molders have already made quality routine. For instance, 77% have a fully implemented quality program; 100% said their program was statistically based; and 85% have a goal of reaching zero defects. Similarly, the survey showed 97.5% on-time delivery.

In Japan, the leading manufacturers have long since stopped shipping any bad parts. Their customers don't tolerate it. These top component producers have elevated quality almost to the level of religion. Product quality is assumed. Delivery is assumed. And a low price--even a declining price over the course of a long-term relationship--is assumed.

What's left to differentiate world-class molders? What will determine how successfully any molder can compete? A consistently high level of product quality may help, but if all competitors offer the same, it won't increase your competitiveness--or more particularly, your profitability.

Overall plant efficiency can have a major impact. Increased efficiency, particularly reduction of indirect labor costs, is founded on common sense and good, timely information. That's where computer-integrated manufacturing (CIM) comes in: making use of real-time information throughout the plant, from order entry to the molding shop floor, to finished-product inventory control.

Some people call this principle Total Quality Management (TQM). As distinguished from merely attaining good end-product quality, TQM is the application of quality principles at every stage of the business, at every level of the company. TQM assumes that process and product quality can be brought under control. What the followers of TQM look for is continual improvement of the entire manufacturing system. TQM means doing things right all along the line, from preparation of a quote, to order entry, toolmaking, maintenance, purchasing, inventory control, production management, quality control, shipping, and customer service.

Historically, the many steps in the manufacturing process have been looked at in isolation, separate from each other and from the ultimate consumer. But each company function serves a subsequent function; the person performing each function is a supplier to someone else along the line, until the molded product reaches the end user. There are nearly always two by-products of doing things right: lower costs and higher productivity.

As the '90s wear on, truly world-class molders will separate themselves from their merely excellent competition by optimizing their entire operations. Machine and process optimization is important, but it is a natural--and easy--step to go further. By optimizing material control (both raw material and finished product) and indirect labor (front office, production management, maintenance, quality control), molders can gain much more in efficiency and productivity.

TABULAR DATA OMITTED

TABULAR DATA OMITTED

A NEW LOOK AT PRODUCTION MONITORING

Early discussions of real-time production monitoring--and the broader concept of CIM--focused on the injection molding machine and the auxiliary equipment immediately surrounding it. The emphasis was on the molded part and the process and process conditions that produced a good part. Likewise, much of the justification initially offered for such equipment also centered on process improvement.

No one can deny the central importance of the molding machine in the production of a molded part, but it is the integration of many disparate functions throughout the plant that results in profitable operations. In fact, as many as half the users of production-monitoring systems do not use their process-optimization capabilities. Yet they get significant benefits from the system in terms of materials planning, production planning, and more. They simply do things better throughout the molding plant.

Figure 1 illustrates the central role played by a CIM production-monitoring system in the management of an injection molding plant. In every corner of TABULAR DATA OMITTED the plant, the system frees up personnel or makes them more efficient; it speeds work up or does it better; it eliminates error and redundancy; it saves time and materials; and it improves management's control and understanding of the manufacturing process.

MONITORING THE PROCESS

What CIM systems did first--monitor the process--they still do best. But production monitoring systems have evolved far beyond the part counters and event monitors of a decade ago. Cycle time, inject time, gate-open time, melt temperature and pressure, hydraulic pressure, mold temperature...if it can be measured, it can be monitored.

The systems can easily monitor auxiliary equipment, too. Part-attribute data like weight, dimensions, and even color, can be entered manually or automatically. Operators can log rejects and the reasons for them. The systems clock downtime, and operators can record the causes.

None of these capabilities is especially new, and most of the better systems on the market can perform them. Two things are critical, however. To be of greatest value, the information must be 1) real-time and 2) available to many users throughout the molding plant.

The value of collecting, analyzing and storing process data--particularly SPC data--has been discussed in great detail before. In the present discussion, it is important to note only that a CIM system can automate the gathering of data, and thus can save considerable labor. At the same time, the amount of data collected can be increased cost-effectively, and more people can get easy access to it, so the information is likely to be more accurate and more useful.

Also worthy to note briefly is CIM's ability to "lock in" the conditions that produce a good part. When this processing "window" is programmed into the real-time monitoring system, an alarm sounds when the process begins to drift and before a bad part is made. Add a robot that actually removes any part produced under questionable conditions, and you have a system that approaches zero-defects manufacturing--all without increasing QC labor.

PRODUCTION PLANNING STARTS WITH ORDER ENTRY

A CIM system should integrate front-office computers with shop-floor production monitors. The initial benefit is the immediate plantwide availability of job information as soon as an order is entered. Foremen, tooling managers, and materials-control and quality-control people all can begin scheduling their work using the same basic information. Machines and tooling can be scheduled simply by entering appropriate run-length, cycle-time, and production-efficiency information into the system. If the job has been run before, the process is even simpler, because all the information will have been stored from previous runs.

Even after production begins, the integration of plant-floor and front-office computers pays off. The system will help track production and inventory of finished parts, along with multiple shipping dates associated with blanket purchase orders. Not only is it easy to keep records of parts produced and parts shipped, but changes in a customer's standard order are quickly accommodated. Parts availability is verified and, if necessary, additional production can be scheduled.

If the job is a rush order that necessitates disrupting work already under way, the job-scheduling software automatically recalculates production schedules for the interrupted job or jobs and flags any major conflicts.

Customer service is generally improved, too. With real-time production information, accurate job schedules, and historical records of process conditions associated with any lot of parts, the molder can easily answer questions about costs, delivery dates, part quality, and much more.

Clearly, the orderly and accurate flow of information from order entry to production control to the shop floor, tool room and so on will boost efficiency dramatically, reducing manpower needs or increasing the productivity of existing personnel.

But more than that, the CIM system allows each function in the sequence to be performed better. Because of the accuracy of the system, jobs can be scheduled more tightly, without "fudge factors." With real-time machine-operating data, production planners can confidently schedule the end of one run and the start of the next.

In addition, because these machine and tool performance data can be collected and stored over months or years, it is possible to know such subtleties as the fact that Mold A can be counted on to run at 94% efficiency in Machine 12, but at only 82% efficiency in Machine 7. Users report significant increases in machine utilization rates and dramatic reductions in labor requirements. This means that more work can be performed without more equipment or personnel; that more maintenance time can be scheduled without reducing output; and that production management has more time to spend on training, long-range planning, or quality improvement.

Finally, just as order-entry information makes its way through the system to increase production efficiency, production data feed back to the front office to increase speed and accuracy in customer billing and general ledger accounting. The circle is closed completely when historical production and cost data are used to prepare more accurate and timely quotes for new work.

MAKING THE MOST OF MAINTENANCE

Maintenance is critically important to sustaining high levels of equipment utilization and product quality. It is also an area where hard information may be difficult to come by and indirect labor costs can be high. A CIM system can help on all counts.

Consider preventive maintenance. With a real-time production monitoring system, some preventive tasks may be put off until a process reading (an inconsistent shot weight, for example) indicates it is actually necessary. When problems occur more suddenly, there's a current record of process readings to help with diagnosis.

Suppose a molding machine appears to be running well, but all of a sudden, parts begin flashing. Without real-time production monitoring, maintenance personnel might make an educated guess as to the problem, make adjustments, and resume molding, only to discover that the problem remains. With production monitoring, a record of inconsistent injection-pressure readings immediately prior to the first bad part might signal a problem with a hydraulic pump, for instance.

It is not unusual for a molder to save several hours of maintenance labor each day when the production-monitoring system is available to aid diagnosis or to schedule maintenance around a running job. (All the same benefits apply to mold maintenance, as well.)

The CIM system should also include a maintenance record that logs all maintenance functions performed into the computer memory. The system can document maintenance activities as they occur, produce forecasts of future preventive-maintenance activities, and generate reports and displays.

Using this information in conjunction with advanced machine-scheduling functions, it is possible to complete maintenance tasks with as little disruption of production as possible. Machine work schedules extending over weeks or months can be programmed to include downtime periods for maintenance. If the unforeseen occurs, the entire production schedule is recalculated automatically when the length of expected downtime is entered. If critical delivery dates are affected, this fact is immediately obvious. Customers can be warned or schedules adjusted to avoid delay of essential parts.

GETTING MATERIAL UNDER CONTROL

According to the latest estimates from the Society of the Plastics Industry (SPI), direct material costs represent 35% of the cost of an injection molded part. When molders run millions of pounds of material per year, it is easy to see that serious money is involved.

There are two general aspects of material control that are addressed by production monitoring and CIM. At the simplest level, a production-monitoring system helps a molder fine-tune material usage. Machine and mold efficiency data, accumulated over time and stored in the CIM system, make it possible to estimate more exactly the material required for a run. By monitoring production and scrap rates in real time, the system provides up-to-the-minute revisions to those estimates so that more material can be purchased or taken out of inventory to ensure no interruption of production.

More sophisticated materials-management data can also be provided by monitoring process and part-attribute data, which could, for example, point the finger at an overpacking condition that would waste costly material and perhaps generate poor quality parts. Sensing a variation in injection time or pack time, or directly measuring part weight on each shot, could shortstop costly material waste, especially on large parts.

With real-time information on key processing parameters or part attributes, the molder knows immediately when the process begins trending toward tolerance limits, even before it begins producing scrap. Correct the problem quickly, and you save material. With better information on production and scrap rates, it's possible to minimize overruns and underruns and the material waste that results.

A broader range of benefits accrue with the integration of shop-floor production-monitoring and front-office manufacturing resources planning (MRP II) systems. Current and scheduled production is analyzed to determine requirements for resin, additives, colorants, inserts, and other materials, along with primary and secondary production equipment, packaging materials, and personnel.

Production quantities are processed through the appropriate bill of materials to determine if required resources are available, to purchase what is not already in inventory, and to determine the real cost of the part. (Availability, allowed levels, and cost impact of regrind can be included.) This same information, updated continually throughout production, then can be used to support accounts payable and receivable, and general ledger calculations such as cost of sales.

Assembly and other secondary operations are also included in process routing and are monitored so that management has an accurate, real-time picture of work-in-process, materials usage, and good and bad parts produced.

TRACEABILITY AND VALIDATION 

For jobs requiring comprehensive lot control, material control extends to the ability to record and track material lot numbers from receiving through quality assurance (if necessary), warehousing, production, shipping, and even end-use of the molded product.

Increasingly, product-liability and quality concerns, along with government regulations, are leading manufacturers to implement parts validation and traceability systems, and to require their use by vendors. As part of a CIM system, these capabilities extend the molder's control over what's produced and ensure the quality and integrity of those products.

The system can be programmed to generate barcode labels that are affixed to bags and boxes of finished parts. The barcodes then are used to associate a given lot of parts with the raw-material, process, and part-quality information stored in the production-monitoring system. If, at any time, a problem is found with a single part or a whole lot, it is a simple matter to scan the barcode on the bag, box or pallet and retrieve all the stored production and quality information.

The molder can find out the job number; the time and date the lot was produced; the number of parts in the lot and the entire order; which mold, machine, and operator produced the lot in question; and the SPC data recorded at the time. Then it is possible to identify and isolate a process problem that may have existed for a short time and determine which parts were made under those conditions. All potentially faulty parts can then be quarantined and the rest of the order released with confidence. The molder doesn't have to scrap an entire order, and the customer has evidence that parts delivered meet quality requirements.

PRODUCT ENGINEERS NEED MANUFACTURING DATA, TOO

Considering the emphasis being given the concept of design for manufacturing, it should not be surprising that the information generated by a CIM system should be used to help product engineers make things better.

In some captive manufacturing plants, for instance, product engineers are being linked directly with the manufacturing floor via production monitoring systems. When the molding plant begins to run a new product, engineers will be able to see how the part is running, whether cycle times are on target, if the critical dimensions meet specs, and even if visual criteria are being met.

Using CIM as a tool to achieve total quality, all that information is immediately available, in detail. Part design, tool design, manufacturing and cost estimates--all can be fine-tuned or changed more quickly, reducing time to market, reducing manufacturing delays, and increasing productivity and profitability.

MANAGEMENT'S WINDOW ONTO THE SHOP FLOOR

A CIM system can be management's window onto the production floor, too. The large volume of data on machine operation, which is accumulated largely automatically without input by production people, can give manufacturing management insight into the ongoing performance of specific machines, particular departments, and even the plant as a whole.

Machine performance history reports list reasons for downtime, the amount of downtime, and the total run time for each machine. Downtime summaries highlight trouble areas. Wasted hours spent waiting for people, material, mold or machine maintenance, and so on are identified quickly. Production management can see which machines are losing productivity, and how much productive time has been lost and why, and then make informed decisions to correct the problems.

Keeping track of production output is equally important. All running jobs can be tracked by mold number, job number, and shift; and information is generated on total parts made, total run time, total downtime, and total parts scrapped. The system can calculate efficiency rates and yield rates automatically. With large-capacity hard-disk memory devices, it is possible to maintain weeks of production data active and readily available in the system.

Much of these same data can be used in cost and contribution calculations that may be executed in the system. Managers can track material and labor costs on a particular job and compare them with standards input into the system. The output is a detailed variance report of profit or loss. Reports can be generated periodically during a job run, and a more detailed job-cost report can be issued for a completed job. Comparing production data on the entire job against standards quoted on the job results in accurate variance figures and detailed profit or loss calculations.

Operator performance can be tracked also, through the use of log-in/log-out procedures. Thus it is possible to associate specific employees with other productivity data gathered in the system. You can compare the results of different operators working on the same job, or a particular operator working on several jobs. Reports can be generated on operator performance over a shift or over longer periods of up to many months. When maintenance, QC or other support personnel also log in and out at a particular machine, management can check response time and relate it to downtime reports accumulated elsewhere in the system.

In an atmosphere of total quality management, continual improvement is essential. To continue to improve, management and employees both need a constant flow of information on how they are performing and how well their equipment is functioning. Information allows them to understand the root causes of problems and to evaluate the efforts made to correct them. A CIM system, consisting of a comprehensive production monitoring system that is integrated throughout the manufacturing company, is not just a useful tool for improving quality and productivity. It is also a highly cost-effective way to gain a competitive edge through total-quality molding.

'It's Kind of a New World'

"When we put in the system six years ago," recalls Connie Kunkel, quality administrator at Teledyne Water Pik in Loveland, Colo., "everybody got very excited, because now they could actually see how each machine was operating in real time." The company in integrating many of its molding and manufacturing operations using a ProHelp computerized production-monitoring system from Mattec Corp. The system gathers data from machine interface units on 22 injection machines and distributes data to six terminals throughout the plant.

What excited people at Teledyne Water Pik was the ability to look at how individual machines were running at any given time and, over a longer time frame, analyze such things as machine productivity and utilization of equipment.

This kind of analysis, Kunkel says, can help you increase utilization of existing machinery in the short term, and in the longer term, no justify addition of new equipment. Largely because of this information, Teledyne Water Pik has increased its equipment utilization from an average of about 65% to almost 90%. Part of this is due to reducing downtime by 2%, accomplished by tracking preventive maintenance more closely.

By getting better control over machine scheduling and increasing setup efficiency, most production runs now cover two weeks' to 30 days' manufacturing requirements, compared with running six-month quantities before. The most obvious benefit here is reduced finished-product inventory.

Ms. Kunkel and her team can also analyze tooling performance. "We may produce the same part on two or three tools, so we can track run hours for each one," she says. "We have documentation that shows which mold runs better in which machine. Our tooling engineers can now more easily justify putting money into tools or repairing existing ones."

Almost everyone in the molding plant has direct access and interaction with the ProHelp system. Supervisors and lead-persons can track daily production and scrap levels; and materials handlers, tooling engineers, and quality inspection teams all have real-time data that can be used to better schedule their work.

The system even works long-distance. Most product assembly takes place in Fort Collins, Colo., about 12 miles away. Says Kunkel, "They were forever calling to find out when a job was going to be run." The two facilities are due to be networked this month, giving 150 people in both plants direct access to monitoring data. To find out when a part run is scheduled, anyone at either facility can simply check the real-time scheduling screen on the CIM system. Product engineers, also located in Fort Collins, will be able to tap into the system to see first-hand how a new part is running, how dimensions or even esthetic factors compare to specifications.

"Now that the system is helping us determine where we have problems," Kunkel concludes, "we can get down to the really hard work. Where previously we may have taken a band-aid approach to problems, now we can take a more diagnostic approach to find the problem and fix it. "It's kind of a new world."