In both QFS and API's custom systems you can expect cost savings, better quality, tighter control of manufacturing processes and shorter time-to-market. Below you will find a description that will help with possible ways to look at a system purchase in terms of ROI.
Recently I had a look at one of API's customers and their use of API's QFS Manufacturing Execution System. I did this to try and understand a variety of the different values QFS offered to them and to have a basis for explaining the system and its value to other enterprises.
I was challenged to this task by a friend who measures improvements to his manufacturing endeavors in terms of two elements: cost and control. As production management that is how he achieves high levels of competitiveness.
My own view looks at the to attributes manufacturers provide along with their products: cost, quality and calendar. The competitive company delivers lower cost, higher quality products in a timely manner.
Neither way of looking at manufacturing enterprises is perfect but looking at the combination of cost, control, quality and calendar can help assess improvements in manufacturing. These improvements can include a software system like QFS, purchase policies, changes in product design, or updating manufacturing techniques. The analysis can assist manufacturers deliver higher quality products, reduce time-to-market, get ISO 9000 certification, increase yield, or any of a variety of other goals.
It was with that background that I had a look at one of a customer's use of the QFS MES. This customer was the first customer of the system and have had it in use for three years. In that time they decreased product cost dramatically. They increased production volume two orders of magnitude. They got ISO 9000 Registered. They implemented flow manufacturing. QFS was a central element of their production strategy.
With QFS as a significant part of their production strategy, they requested and were part of the development of many parts of the system. API spent a significant amount of time working with this customer and has intimate knowledge of their entire operations. While they do not seek publicity this close relationship made a great opportunity to look QFS from a customer's point of view.
My friend was right. This analysis from the cost, calendar, quality and control perspective was instructive to me. I hope it is also instructive to you. It is if you see parallels in this analysis and your operations. It may have added value if you find ways the QFS Product Family can help you.
The Quality Feedback System (QFS) is a software system to support manufacturing organizations. While it has interfaces to business systems, it goal is to improve shop-floor related activities. This means collecting data about the manufacturing process without getting in the way. It means increasing yield. It means supporting increased manufacturing volume. It means supporting both high-volume and lot-size-one manufacturing. It is used wherever direct material contact occurs.
The parent of the family is QFS itself. This is a manufacturing execution system (MES). All the other elements of the QFS Family are either portions of QFS, separated to provide scaleable partial functionality, or are designed to interface directly with it. As the parent of the family it is helpful to know what QFS is and how the MES relates to the overall manufacturing environment.
On one hand, QFS is a collection of manufacturing systems in the manufacturing process where the stand alone functions perform some task in the enterprise. While a detailed functional description of the QFS MES can be found elsewhere, the various elements of the system include
QFS is also a horizontal integration of those elements and islands of automation, tying together disparate elements of the shop floor to create a unified manufacturing environment.
The horizontal integration of QFS allows these islands of automation to work together in a coordinated fashion. The left hand then knows what the right is up to. Shipping knows, automatically and for free, that the products going out the door have been tested. Final test knows that intermediate tests have not only been done, but can also check the results. Managers can see a view of all or parts of the manufacturing process. Engineering can look at detailed data for analysis, prediction and assessment. An important note about the horizontal integration of systems is that it extends beyond facility walls. Horizontal integration includes outside producers (OSP) of assemblies and components into a single, virtual enterprise.
In addition, QFS is a vertical integration of the shop floor with other systems in the enterprise including the planning, finance, and design systems. This allows the BOM that the planning system uses to be accessible to the test systems so that data that finance uses will match data that production uses.
QFS has direct ties to the manufacturing resource planning (MRP) system of the enterprise. The ties to the planning system make the initiation of production well coordinated with demand. It also improves inventory reporting to that system to increase accuracy and timeliness.
The inventory movement and material backflushes to happen automatically as production occurs, without error and without taking time out to make a separate transaction in another system. It therefore improves the quality of the production planning process.
Likewise, revision control on the shop floor is improved by the coordinated distribution of new design and revision data to the entire shop. The same BOM that the planning system is using is available to the test systems, increasing data accuracy.
Again, functional and system definitions of QFS can be found elsewhere. If you are unfamiliar with QFS and the functions of an MES finding these definitions may be a good idea. Rather than go into specifications of the system, the state-of-the-art tools it uses for analysis and production modeling I return to the four-element analysis.
Those four elements of manufacturing that QFS improves are cost, control, quality, and calendar.
Cost in manufacturing comes in many forms. The primary elements that increase costs are:
Each manufacturing enterprise has a different balance of these elements. A systematic approach to manufacturing can reduce the costs from each of these elements. QFS reduces the cost from all of these elements.
The degree of the savings is very dependent on the manufacturer involved, the systems already in place, and the commitment of the company to achieve the savings. Again, these are the savings that I saw at one customer site. Hopefully you can see a relationship between these savings and those at your enterprise.
Materials costs as addressed here will include only direct materials savings. Indirect materials costs, such as the interest costs associated with excess inventory, will be dealt with elsewhere. With that said, how does an integrated manufacturing system reduce the direct materials cost?
One simple way QFS reduces materials costs is its high-resolution inventory function. The inventory system in QFS is complete, supports the financial and planning inventory system, and spans the enterprise. This allows materials to be tracked and used effectively. It reduces the situations where inventory cannot be found until it is obsolete. This situation occurs more and more frequently as product life cycles are reduced and product mix is increased on the shop floor. Yes, this is also a control issue.
Another control issue that directly effects cost is the distribution of product revision changes. As product and revision mix occurs on the shop floor, occasionally incorrect builds are made. QFS supplies a mechanism to distribute current product definitions accurately and effectively.
One direct way that QFS reduces materials cost is through its label generation and printing functions. This means that only label stock inventory needs to be carried, not actually the labels themselves. It also means reduced scrap as labels are printed on demand on the shop floor. No more tossing pre-printed labels because of changes in revision and requirements. (It also allows the flexibility of customer-specific and custom labels with no increased materials cost).
QFS prints the labels for product themselves including serial numbers, barcode and others. It also prints packaging labels, shipping labels, receiving and material handling labels, and a host of other types. The data that is printed on all of these labels are pulled from the central QFS database and represents the most up-to-date information at the time of manufacture.
An often unexpected benefit of QFS doing label printing is the reduction of the number and types of labels. When it is easy to see the numbers and types of labels that are used it then becomes possible to reduce redundancies and consolidate materials.
By building the right stuff excess and obsolete is reduced.
When it exists excess and obsolete is more easily used by improved aliasing.
If excess and obsolete can't be used, selloff is easier as more information is available to the selling agent about what is being sold, its condition, and where it is.
Aliasing of materials is supported in the system to allow materials substitution without unnecessary redefinitions of product BOMs and hierarchies.
Another way that QFS reduces materials costs is making data and analysis available for design engineers review materials performance in production test and the return and repair cycles. This not only helps improve design for manufacturablility and increases product reliability but also allows designers to build products with lower cost and common parts. It is an effective supplement to the tools provided in the design systems.
In addition, materials (and other costs) associated with rework are reduced with QFS. How QFS reduces rework will be address later.
QFS reduces labor in some obvious and some suprising ways.
QFS reduces two materials-related labor costs. The first is the labor associated with receiving, receiving inspection, and materials stocking. The second is cycle counting of materials.
In the case of receiving and receiving inspection, not only is the inspection and data entry process streamlined, a history of vendors and parts is made. This allows skip-lot inspection for trusted supplies.
In the case of cycle counting, calculations of materials due for count are automatic. The counts themselves are simplified with portable terminals. Material reconciliation is done with authorization levels and automatic calculations of dollar levels.
QFS also reduces the cost of procedural documentation that needs to be delivered to the shop floor. It does this by using training data and carrying training information to the point of use on the shop floor. This reduces ISO 9000 procedural documentation requirements.
QFS reduces the cost of data collection by standardizing the data entry interface, by making much data entry implicit rather than explicit, and by checking data integrity and validity on-line, while the questions are relevant. Implicit data entry means not only the automatic collection of data by systems that in other environments just throw it away, it also means exhaustive data is collected easily.
By enhancing shop-floor analysis capabilities QFS reduces direct labor costs. This includes a variety of things such as early detection of problems and detection of subtle problems. It also includes knowing the extent of known problems to better plan for recovery.
While QFS does not have explicit planning functions, it supports the planning process in a variety of ways.
Kanban flow manufacturing reduces materials on the shop-floor and reduces the finance charges associated with materials. Flow manufacturing is not required for QFS but it is one of the fundamental modes of the system.
Materials substitutions between work orders are also allowed as materials are available for cross-lot and cross-unit use. Engineering analysis of possible substitutions is supported as well as improved knowledge of material availability.
Better tracking of materials increases the accuracy of data available to the planning functions and increases the accuracy of the resultant plan.
It isn't immediately obvious that QFS reduces overhead, however, it is the case.
QFS empowers lower skilled workers to do more and thus reduces the burden on engineering for analysis. QFS allows every level of user to do more analysis than previously possible. This includes not just direct shop-floor labor but also indirect engineering and management.
The computer hardware that is normally used for other functions, like test and inspection, becomes data-entry points for the system. In addition, business system terminals are reduced as QFS performs many of those interactions automatically.
The Test Database portion of QFS reduces the amount of effort test engineering must spend on data gathering and data analysis. These functions are all supported in QFS. In addition, the protocol used is meant to minimize the load on test engineering for integration. By using this protocol, test engineering can focus on test and test analysis, not on IS issues.
Control comes down to knowing what you are doing, when you are doing it, and why you are doing it.
It seems to be such a simple question. Sometimes the answer is not so simple.
Better training for people on the shop floor is ensured by validating training levels at the point-of-use. The training matrix is available to all QFS Clients. It tells them if they are due for training or not. It can be used to either warn people about new training requirements or it can stop activities altogether if training is up to date.
Managers maintain the training matrix for operators. In addition to the point-of-use training verification, the training reports are used to verify cross-training, even for operators who are not currently engaged in certain activities.
This control issue also is a cost-effective way to conform to ISO 9000 requirements when compared to the costs of distributing up-to-date procedural documentation.
QFS ensures that no non-functional product goes out door with shipping and packaging quality data checking. This verification process for critical activities can be done throughout the manufacturing process. Elimination of shipment of bad product is assured.
Standardized defect lists and data integrity checking at the time of data entry means that manufacturing controls are done using validated data.
Centralized data uploading and reporting means that inspection, test, and measurement data is unified and gives a better and more consistent view of the entire manufacturing process. There are 1000 ways control can improve when the information that control is based on is accurate and up-to-date.
QFS supports better accountability of outside vendors with three mechanisms.
One is using the receiving inspection system. This builds a history of a vendors quality rating by associated the test and inspection data with the vendor. In the case of trusted vendors skip-lot inspection is supported to reduce the costs of that accountability.
OSP houses can provide quality and build data on diskette that accompanies shipments to support the internal receiving processes.
By supporting remote QFS Clients in OSP houses, build data for complex assemblies is on-line in the central QFS System. This feature of remote clients has been used locally and internationally. It can be used for remote production, test as well as remote inspection by off-site inspectors.
QFS supports easy and accurate cycle counting. It has a mechanism for user authorization of cycle counting charge-offs, which is dollar-based. This means that misplaced material is tracked down be appropriate people.
Doing the right thing at the right time assures timely deliveries of the most up-to-date product with minimal inventory levels. QFS supports this in a variety of ways.
QFS has controls so that only valid products of current revision are built. The data comes directly out of the MRP system. The data is centrally stored and is distributed in a unified fashion to the entire shop floor.
Centralized data downloading for every system means tests, inspections, labeling, production and shipping is all done with the latest and most uniform data. Revision changes are apparent immediately.
Sales order fulfillment matches valid product revision levels. This helps when revision changes have happened, when customer and engineering shipping holds exist, and when old stock is being cleared.
QFS has the capacity for sales order sequencing, build to forecast, build to manufacturing order. In minimal inventory, flow manufacturing environments sales order sequencing usurps some of the planning function in a steady-state process. While it does not work in all environments, it is handy when it does.
Knowing why manufacturing is doing what it is doing is often out of the hands of the shop-floor staff. Even so, communications to other departments can help the shop floor know it is doing the right thing. QFS supports that communications.
Link from the MRP planning system allows people on the shop floor to verify demand as product builds commence. In environments where demand fluctuates, visibility into up-to-date demand can prevent unintentional builds and premature inventory.
Sales staff can track orders through the production process in the same way that manufacturing management can. This gives them a way to work with customers for planning and delivery.
Data can be supplied to product marketing and sales to facilitate both long- and short-term product and sales planning.
A core aspect of QFS is its ability to integrate quality information into the manufacturing process. It is, after all, the Quality Feedback System. It distributes what needs to be done and how it needs to be done. It records the results of what was done. It provides the analysis tools to see how well it was done. It improves quality.
To distribute what needs to be done QFS holds a relational model of product descriptions, sales orders, customer information, vendor information, test requirements, production routes and a large host of others. All of this information is available to every place on the shop floor where it is needed. To make the information effective, it is hidden when it is not needed.
To record information all the production processes are on-line, interacting with the central QFS system as operators, technicians and engineers perform normal activities. The degree of connectivity is transparent to them. They do what they need to do and the system guides it and captures it.
Then there is the analysis. Users at any level can analyze the production process from any perspective desired, from individual assemblies to entire product lines. From one station to the whole shop floor. From first-pass yield to Nth-pass yield to aggregate yield. From a component to the whole assembly. If you tear something apart and rebuild it in a different way it can tell you what happened to all the pieces. So, what individual assemblies were shipped between 3 and 5 PM on Friday at the end of the last quarter that were handled by operators with brown eyes?
The fundamental point is, if you know what's going on you can make it better. If asked how well you did, you can tell someone.
No manufacturing system is worth anything if the data does not go in. QFS makes that easy. Users get help from it to do their jobs and in the process the data is collected.
The standard QFS Manual Client is a LabVIEW-based client that talks the full spectrum of the QFS protocol. It has an easy to use, color guided, data-driven optimization of the information presented to users with data validation and pull down menus.
It is easy to use. Because of that the Manual Client encourages putting data into the system. The data that goes into the system is standardized and validated. All the good data goes in and none of the bad.
Test systems have an easy-to-use systems interface, and makes the same thing true for test systems, via test system developers, as is true for end users using the Manual Client. By making it easy to add data to the system, data is actually entered.
Data integrity is likewise improved as the test systems themselves have the data they are storing standardized and validated.
QFS has a variety of reporting tools. These include standardized reports for often-used production analysis. It also includes ad-hoc reporting tools to enable custom reports for specialized purposes.
QFS has unified views of data and feedback from the entire production cycle. For example, yields can be looked at for the collection of production, on a area-by-area basis, or by process type. This type of comparison finds strengths and weaknesses in the manufacturing process. It also allows root causes to be found.
The standard QFS Manual Client has an easy-to-use graphical user interface which encourages users on the shop floor to look at enterprise data while doing production. This includes data like current product revisions and specifications.
The Manual Client also includes easy ways to view production results. Users can see what the trends are in there work area. This enables more effective quality management by identifying problems in a way that allows them to be addressed.
Shop-floor managers have the same access and can use current data to support staff meetings where the real and current problems are addressed. The number one problems can be addressed and current yield information allows meaningful discussions on status and improvement with production staff. These same tools allow engineers to review production to improve processes, designs, and staffing.
Summary reports can be used by both engineering and management to get snapshots of production volume, yield, shipments and other things. QFS even supplies data the end-customers my require about production practices and quality details related to shipments.
Data accuracy is improved. Views into the data are improved. This improves manufacturing accuracy. This improves quality.
QFS supports a number of manufacturing paradigms and manufacturers can pick the manufacturing method that allows best throughput, time-to-market, and varied product lifecycles.
Most of QFS' support of increased production throughput is indirect: it supports better planning and control, elimination of wasted effort and more efficient use of resources.
QFS reduces the time to collect the data used to improve production. It includes the optimized QFS Manual Client for data collection. This client presents operators with menus that allow data to be stored (and analyzed) with little in-process effort.
Data that is collected by automatic Test Systems requires no effort by the operators at all. The operator performs the tests that they normally do. Data collection is transparent and complete.
For shipping, cycle counting and stock room functions, use of the QFS Radio Frequency (RF) terminals means that clerks go directly to do the job and carry all the information they need with them The time wasted gathering and collating information is eliminated.
QFS supports the analysis and design of more time-efficient manufacturing. Analysis of time-in-station permits accurate process time analysis. Quality information allows improvement of processes and ultimately the elimination of processes. QFS allowed the complete elimination of production final inspection by improving quality to such a degree that the inspection was no longer necessary.
Kanban, flow, lot, and batch manufacturing are all supported, even in parallel, to allow the manufacturing that increases throughput.
Flexible routings allow manufacturing resources to be deployed to work around manufacturing bottlenecks. In addition, QFS allows stations to be dynamically reassigned and used in parallel to add capacity to bottleneck operations.
Data availability allows fast MRP systems to do "what if" analysis with direct, timely, and valid data. This data includes the actual production values rather then estimates.
Distributed systems allow outsource producers to offload internal work while remaining an integral part of the manufacturing process. This integration includes collecting data from files supplied by outsource producers.
It also includes direct connections between the outside producers line and the QFS Server. The QFS Protocol supports these remote links. The links have been used locally and internationally. Test, inspection, and production functions have all been used successfully.
QFS allows the physical mixing of products in the production process while keeping the data separate. This allows the flexibility in production that reduces time to market and allows shortened production runs for short-lifecycle products.
Tracking of mixed manufacturing and mixed lots allows reduced time-to-market for new products and new revisions of existing products. Builds of many revisions of a product can be done at once with separations of the instructions, descriptions and data collected.
Engineering builds can be done in conjunction with live production with no compromises in data integrity. Reporting for primary production is easily separated from the engineering data. Data rollups are also allows to show the proportional impact of engineering builds on production.
Automatic distribution of new product information to all areas of production allows building many revisions of products to be built at the same time. This does not compromise the control that is required as production starts.
If a revision change requires the conversion of one revision to another QFS supports finding the old revision in the process stream. It then allows a conversion of the product from one revision to another without compromising any of the data. A complete history of the changes is kept.
This cost, control, quality and calendar analysis is only one way to look at the effectiveness of a system. In this analysis much of the interaction between the elements has been ignored. On this day I have used this particular analysis to look at the system. While others can be applied, I hope this one inspires potential users of the system to have a closer look at it and members of the QFS Product Family.
At API we have tried to make QFS the most effective system possible from every perspective. Judging from the reactions we have from customers of the system we have been successful. Our success helps them be successful as manufacturers and businesses.