Regardless of what you are building or how you are building it, whether you are artfully removing material to make a dugout-canoe, or methodically building up layers of veneer to laminate a rowing shell, the product you are producing is a mixture of information and material, and most of the value of the finished work is added along with the information. The information you are providing is the basis of your craft. What you are selling is your craft -- it is what you know how to do.
Because you have a relatively short amount of time to spend practicing your craft and because the cash value of each sale is relatively small compared to the cost of raising a family, you need to sell this information repeatedly and often. To increase the frequency of these transactions, you have figured out how to translate what you do into a "process". The process is a series of steps that can be performed sequentially and ultimately taught to an employee (or employees) who can "amplify" your effectiveness, producing more goods for sale than you could ever produce alone. This is the basis of the factory and the basis of manufacture.
When you establish the factory, you establish the "process" at a scale that requires a team of employees to produce the product you used to manufacture alone. This serves two functions: 1) it provides an amplification factor, allowing much higher productivity, and 2) it "balkanizes" the information. By "balkanizing" or breaking the craft into small pieces, you avoid the pitfall of training each of your employees to become your most dangerous competitor.
The change in scale from craft to manufacture turned you from a craftsman into a manager. Once you were a craftsman. Now you hire and train craftsmen.
As a manager, you have been forced to learn a new craft: namely, management. Your responsibility is increased. You undertook the initial expansion of your craft to feed your family. Now you have to consider the families of your employees as well, so you have learned to increase your manufacturing capacity further.
You made decisions to buy specialized machinery to reduce production bottlenecks.
You learned to "batch" activities that require substantial setup time, creating inventories of parts which are subsequently depleted as the parts are assembled into products.
You learned to refine the design of every part of every product you build, to eliminate awkward or unnecessary operations.
You learned to juggle -- to schedule cash-flow and operations, attempting to keep too much from happening at once.
Both specialized equipment and batch processing increase your operational efficiency, but they also increase the complexity of your operation. Batch processing creates two major problems: 1) coordinating the production rates of unrelated processes, and 2) scheduling the replenishment of inventories that must be created by your own production or acquired from outside vendors.
Specialized equipment is expensive and it requires costly tooling and painstaking setup, reducing the flexibility of your operation and making incremental design changes difficult.
Both these solutions tend to alienate your employees, forcing them into ever more routine work patterns and making them feel like mere extensions of the machinery they are using, rather than craftsmen. This can lead to quality control problems requiring rework or returns.
The quest to eliminate problems before they occur has led managers into ever more analytical manufacturing schemes aimed at maintaining "quality" by reducing the probability of the problems occurring. The fundamental premise of quality control is that it is cheaper to prevent an error (almost regardless of cost) than it is to deal with the consequences of shipping one or reworking parts or finished goods to eliminate defects after they occur.
This is an extremely important concept, which must be implemented at all levels of your organization simultaneously. It is normally accomplished through incremental redesign of both the parts and the manufacturing processes that produce them through the implementation of quality assurance procedures (such as multiple levels of inspection and card / tag systems, or the more rigorous Kaizen approach brought to us by the Japanese).
Over the course of a few decades, the way products are designed, manufactured and marketed has changed dramatically. There are many factors working in concert to force this change, including diminishing resource supplies, increasing population, increased number of producing markets, and the development of easy-to-use enabling software which allows information to be organized, and allows this organized information to be utilized more effectively.
Over the past 10 years, access to information has changed the way most business is done in this country and in the world. The proliferation and increase in capabilities of cheap desktop computers has made this change possible. In many ways, the experience gained in the introduction and development of this technology, and the explosive and ill-coordinated growth of the industry which provides these machines, has provided new management tools which will help the rest of society adapt to the change.
We are now on the threshold of a major change in the way things are manufactured. The gap between the cost of manufacturing a customized product and a rigidly batched product is disappearing rapidly.
"Leading edge" manufacturers are already offering the customer a seat at the design table. To date, the available range of "manufactured on demand" products is limited. Most are relatively simple variations of well-defined products of traditional design, such as "made to measure" blue-jeans for fashion-conscious women. But one can imagine that more complex products are not far behind.
Circular stairways, custom fitted bicycle frames (cut and assembled to the customer's personal dimensions), or acoustic guitars "voiced" to the customer's individual taste seem to be excellent fits for this technology.
The closing of the gap between design and production that this technology allows will have a long term significance equal in magnitude to the introduction of electronic typesetting, which within a period of 5 years turned the entire printing industry upside down.
The entire printing industry based on moveable type had evolved smoothly and incrementally since the 15th century when Guttenburg introduced moveable type. It collapsed in the face of electronic typesetting. In the wake of the introduction of electronic typesetting, machinery, skills and techniques worth tens of billions of dollars became obsolete over the period of less than a year.
The following subsection on basic small-business management activities and decision making tools is provided as a "refresher" to insure that everyone reading the material in this manual has access to the same basic understanding and experience on which I base the rest of this manual.
It is necessarily presented in a sketchy and general way because there is a great deal of material to cover. This manual is about getting your business set up to take advantage of new technology that is becoming available, not about getting your business plan written and your business set up in the first place.
In The Business of Woodwork, Norlin says of the three variables you can offer the customer (price, quality and service), "Pick service and quality, and raise the price to fit. You can only offer all three if you are going out of business." This is a valid position for a custom cabinet shop, but may miss the mark for a manufacturer. In order to compete, you will probably have to change the way you do things in order to allow you to lower your price.
The organization of some of the following material is adapted loosely from Norlin's method. However, the content is based on nearly 30 years of personal experience building a wide range of products, beginning with telephone hardware and parts for jet airliners and ending with cowlings for lighter-than-air blimps. I have built sailboats and sound-systems, sea-kayaks and even a carbon-graphite grand piano, mostly to fixed price contracts. I have broken many of these rules again and again, and when I did, I usually barely lived to regret it.
My manufacturing experience was followed by 3 years developing and publishing computer software for scientific visualization, and starting an audio electronics manufacturing company. This was followed by free-lance writing and a brief return to product design and prototype manufacture.
My most recent contracting experience has focused on organizing information and building information systems for local government.
Managing means, at the very least, always knowing what is going on, when things are scheduled to happen, which activities are off track, and why.
The information you need to manage a business is
almost the same information you need to specify in a contract.
There is a traditional separation of business activities into phases: Selling, Fulfilling, & Reviewing. Each phase then follows similar stages:
I used to explain to my customers in the sailboat shop that before we could get down to the business of getting from what they thought they wanted to what I knew they needed, there were only three important factors we had to discuss:
They were told that all three variables were inter-dependent and that, at best, they could only have any semblance of control of only two at a time.
Any operation has an optimum pace, a natural rate at which things move smoothly. Attempts to increase this pace usually result in unexpected complications. Thus, for example, reducing the available time for a project was likely to increase the cost and reduce the quality.
The customer will ask for all three or perhaps even insist on all three. But that does not make it a valid request. Ten year olds ask for a lot, too.
These are the basic criteria for evaluating an opportunity to bid:
do you need the order?
When shaving markup to get the job, do it on a small job not a big one. On all jobs, consider the management time requirements as well as the potential revenues.
The conservative view that keeps people from moving into new areas reads: "Big jobs provide big headaches and require more time consuming paperwork, bid submittals, etc. Steady work from repeat customers can run as high as 30% of the total sales. A huge job could jeopardize the normal work-flow. Loyalty is required from us to warrant loyalty from them."
All this is true, but the amount of work available from your regular customers may never be enough to warrant (or finance) building up your business into a truly effective organization. Even if the stated goal of your business is to achieve a "steady state" that allows you to live comfortably, rather than a continuous state of expansion, the grim truth is that in any business there is a certain scale of management infrastructure that your business must develop before you can play against the big guys.
As your banker ought to tell you regardless of the field you are in, the rules that accompany this view are sound and are borne out by countless bankruptcies. They are based on the valid assumption that the tools you need to handle big jobs are not developed doing small jobs and that there ought to be reasonable limits placed on your reach until more of the management tools are in place.
The reasons for these rules are obvious. There are real and essentially absolute limits to your management capability, your production capability, and your financial capability. Big jobs concentrate risk, whereas batches of small jobs distribute the risk.
You can ride out slow pay or even no pay on a few small jobs if you have others that are going OK. You may not survive slow or no pay on the big one.
The big guy you are trying to get as a customer probably knows the game better than you do, and if you allow it, can get you into a position where you have a note due (with an enormous amount of interest at stake) and its waiting on his payment. This gives him unfair leverage in a renegotiation. My experience is that the big guys are always the hardest to collect from and always pay the latest. You can't afford to serve as the bank for the big guys and you can't afford to get leveraged into working for free.
An example from the outside world is appropriate here. This is the worst kind of business catastrophe I know of, and it happens again and again. I think Sears & Roebuck invented it. Sears used to specialize in finding innovative products made by small firms and "capturing" them. This was a brutal "hostage taking" strategy (similar to the one now used on whole communities of consumers by Safeway, Costco and Wal*Mart). They found the brilliant product and placed an order with reasonably generous terms for a volume that was well beyond the limits allowed by the rules provided above. The supplier was ecstatic about getting into the Sears catalog and invested heavily tooling up to produce the goods. Sears then renewed the order with terms that demanded that the supplier turn away other work. Things looked rosy. Then, at the end of that contract when the supplier was truly a hostage, Sears offered much more painful terms.
REI adopted the same strategy toward some of its suppliers when it took on Sears' management model. Watch yourself. This problem is covered again in more detail in the next section.
The key questions to ask yourself are:
Rules to apply:
This is a matter of detail. You must consider drive time, problems of coordination, etc., etc... How many truck loads will this be, what is the real cost of freight, etc...
the bid list.
There are two lists: 1) the firms who are you bidding to, and 2) the firms you are bidding against. You need to know who is on both lists. Bidding is expensive, even after you get information collection automated. If the competition is made up of disreputable low bidders (i.e. firms not playing the same game you are, not insured, not bonded, not paying taxes, etc.), they are not your real competition. Why bother to bid against them?
your relationship with the designer.
Review the track records of the all the players. Is the contractor (designer, architect, etc...) decisive and competent, or a flakey, wishy washy, pain in the butt? Do you really need this job? How can you figure this out?
the project manager / owner.
Call around and assess the hassle factor carefully. Generally, people with a lot more money than you have are going to be hard to deal with. People who do not share your language, values, or experience may prove hard to please and even harder to work with.
the quality of plans and specifications
Lots of addenda or revisions evident in the plans indicates continued confusion in the project. One of the primary purposes of money is to reduce stress and unpleasantness. If you can afford to avoid confusion, choose to avoid it.
After you have used the tools listed above and made the decision to compete for the job, the next step is calculation of the anticipated cost of building it.
The need for accuracy
Accuracy is very important because we tend to use multiplication to calculate costs and derive expectations. Everyone uses a very simplistic set of multipliers to estimate cost and time requirements. In my fiberglass boat business, I used area, thickness and specific gravity of the finished laminate to estimate material requirements and cost. Usually you can get close enough to know if a project is worth pursuing without knowing any numbers closer than 15%. Often you can get within 15% on the back of an envelope. However, this does not mean that estimates to within 15% accuracy suffice for managing your business.
Multiplication allows the dilution of accuracy. If we take a number in which we have a great deal of confidence, like 27 sheets of plywood in the rack, and multiply it times a number in which we should have very little confidence, like the $1.20 per square foot we paid for it in 1993, we get a number that we should have some reservations about. If we use the dollar value of our inventory for tax purposes and then use it again in our estimation of the material requirements for a project, and the price has changed to $1.34 per square foot, we may run short on material.
Precision versus accuracy
Precision is not the same as accuracy. Precision is paradoxical, since it appears to increase as we multiply and divide numbers of limited accuracy. A box of parts is estimated to cost $20 and to contain 6 parts. Simple division of 20 by 6 leads us to believe that we know what each part cost to the penny. We do not, and it's probably not worth finding out.
Norlin advises: "Inventory based on what was ordered, do not waste resources on re-inventorying material". This means: keep track of shipments and withdrawals. Establish realistic reorder points.
There is an enormous difference between an organization and a collection of people doing related things in the same building. In order to become an organization, many things must be done which do not relate directly to the tasks people are doing. "Caring" will never be part of an employee's job description. And yet something about the way the collection of people interact must provide them with basic elements of security, if it is to become an effective organization.
What do people need? People need clarity! They need to know what is expected of them. They need:
in a nutshell, they need an "operations manual".
The OM codifies what we are doing
If we don't start writing the manual, we continue to ask and answer the same questions over and over again.
The OM provides critical resources:
Alvin Toffler recently pointed out that North America's virtual domination of the world markets for manufactured goods between 1950 and the mid-1970's did not result from the superiority of our economic system, our manufacturing methods, or our management model. In fact, he pointed out, none of that was true. Our success was a result of our total monopoly on those markets. We had no competition whatsoever.
The Second World War had left this country with a very effective manufacturing infrastructure (based on the war effort having trained hundreds of thousands of previously unskilled workers in precision manufacture). The pre-war British manufacturing machinery had been bombed to oblivion by German rockets and the manufacturing capacity of Germany and Japan had been bombed to oblivion by the US. By mid-century, there was no one in a position to compete with the US. When the US auto industry was driven to the brink of collapse by competition from Japan and northern Europe, the US established import quota systems which effectively forced the Japanese car-makers to set up manufacturing plants in North America. What happened at these factories quickly dispelled much of the mythology surrounding management.
Japanese managers were confronted with the larger American management structure, and American workers and suppliers were confronted with an entirely foreign inventory management model. The Japanese system did not mate perfectly with in-place American corporate / management systems and the net effect was revolutionary. This revolution may have been long over-due, both in the US and in Japan.
In the US, Japanese Kanban "pull" production-flow control systems were installed, "just in time" (JIT) partnerships between suppliers and manufacturers were developed, and the Japanese management model of Kaizen or "continuous improvement" was applied. Employee input at all levels of the enterprise was focused on reducing production costs by identifying unnecessary steps and correcting sources of defects at every stage of the operation. In Japan, the cumulative undesirable effects of some of these management strategies are becoming obvious in the form of "just in time" delivery traffic overloading existing road systems, and related problems.
A new paradigm, known as Flexible Manufacturing Systems (FMS: discussed in Section 6) is beginning to change the way business is done on both sides of the Pacific. We are probably on the threshold of a new level of productivity.
Underlying the development of this Design Manual is the recognition that many of the procedures discovered through examination of the flow of information and materials within large organizations and manufacturing operations are scaleable and thus are applicable to operations large and small. Even the latest CAD-CAM and "virtual-factory" technology is scaleable and can find application in smaller operations.
The most powerful integration of computers and design is found in the Virtual Factory. The term Virtual has been lifted from science fiction and refined through application of the disciplines of Virtual Reality, pioneered by Jason Laniar and now carried on by others. Most of the proposed applications of Virtual Reality have focused on its potential for entertainment. In the Virtual Factory, "virtual" means that the tools and the products they create exist only in the "virtual" world inside the computer, rather than in the "real" world. Virtual factories use computer simulations of the machinery, processes and procedures used in a real factory to produce virtual products which are computer simulations of real products. See further discussion of Simulation in Chapter 6.
Which steps add value? Which steps do not? Increasing the speed of processing material, or even increasing operational efficiency in general, is rarely the most effective way to increase profitability in an operation. Material passing through an operation or a machine on its way to becoming a product or a part of a product gains value only if it is changed during the operation. Value is only earned as the material is actually transformed into completed goods.
We initially described your "craft" (the basis of your business) as adding information to material. Reshaping, removing or attaching material is the primary activity that adds information to the raw material and thus adds value to it.
How much material you remove, how you remove it, what you do with the scrap, and other similar manufacturing decisions all affect your company's profitability, but for the most part are of little concern to your customer.
The key to profitability and competitiveness is providing a workplace that maximizes the percentage of your employees' time that is spent transforming raw material into finished products.
Each operation involved in the production of an item has a total cost which can be calculated. Some necessary steps in the process are non-transforming. They add to the product's cost, but do not add to its value. Materials handling, machine setup, measuring and re-measuring, inventory tracking, maintenance, and defect rework are all examples of cost-adding operations that do not add value. If accurate production information is available, the proportion of your total manufacturing cost that actually earns value can be ascertained.
A general pattern is obvious. The proportion of the cost of manufacture that does not earn value is highest in small general-purpose operations, and falls rapidly as the operation becomes better organized and more specialized.
Elimination of errors before they happen
In the Japanese model, an important function of kaizen (continuous improvement) is to eliminate the need for non-transforming operations by creating a production system that does not allow defects to propagate.
In your own operation, this is most often accomplished through the creation of a production system based on "hard" tooling. Redundant measuring operations are eliminated through the use of jigs and fixtures which control the removal or attachment of material or the size or location of attributes.
The initial cost of constructing this tooling is large compared to the cost of measuring, but the cost is amortized over many cycles, and many measuring operations are eliminated. The application of well designed jigs and tooling speeds the operation and reduces the likelihood of error and the need for rework. This 19th century idea still provides the economic basis for most manufacturing operations today.
Another extremely cost-effective method which substantially reduces the likelihood of error (and adds information at each point in the production process) is labeling parts as they are produced and marking them at the completion of each process. This can be done with cards, tags, or stick-on labels. Bar-code on removable stick-on labels is rapidly becoming the tool of choice for part labeling. In a manufacturing environment where there are several (or many) similar-appearing parts and a possibility that they can be confused for one another in the course of assembly, part labeling systems are nearly as important as jigs and fixtures.
One of the most important aspects of the Design Process is the development of accurate production cost information. It is very difficult to make good decisions without accurate cost information.
The discussion that follows presumes that you have assembled a Design Team which includes most if not all of your production employees, and that everyone has really "bought-in" to the idea that improving the efficiency and profitability of the operation increases rather than decreases job security. The issue is that you are tracking the path of material from place to place and activity to activity, not that you are spying on people or looking for slackers.
An ideal way to determine the actual production activities and costs (and especially the ratio of transforming to non-transforming activity in your operation) is time lapse photography. You could mount a video camera on the ceiling and set it to record a second or two every half minute. The goal is to compress an 8-hour day into about 30 minutes in a way that would still allow you to follow the path of any individual part as it makes its way from raw material to part of a finished product. A more traditional technique that collects a lot less information is based on giving someone a stop-watch and a form to fill out for each part.
Once you have a map for each part, you need to put
together a time-line for it. This kind of work is normally done
using a time management program like Microsoft Project or Primavera
(discussed in more detail later in this section), but there is
no reason to not do it just because you do not have the software,
or to assume that using software will make the process easier
or more accurate. You could certainly use either graph-paper or
Once you have developed a tool like this for every part you make, you will be well prepared to dive into the process of revising work flow and defining equipment purchase priorities. Until you do it, the real benefits of project management will elude you.
Much of Chapter 5 is devoted to developing and applying business Metrics. The concept is introduced here without a great deal of development. A metric is a tool to allow measurement / quantification / evaluation of things not necessarily numerical. A product, part or design specification is based on assigning Values to Metrics.
If production cost (the sum of the cost of value-earning and non-value-earning operations) is accepted as the appropriate metric to guide production, and if "continuous improvements" are made in the manufacturing process to increase the proportion of value-earning operations or reduce the need for non-value earning operations, then processes will become efficient. The cost to produce individual items will fall in almost direct relation to increases in batch size.
Operational efficiency has often been viewed as synonymous with large batch production. However, large batches create substantial problems in other areas (inventory costs are high, and there is always the risk that design changes will be required, demanding rework of inventoried items). For this reason a balance must be developed between efficiency of manufacturing (which is minimized when batch size is infinite) and inventory cost (which is minimized when batch size = 1). This balance is evaluated based on a different metric.
The appropriate metric is based on three factors: throughput, inventory expense and operational efficiency.
Throughput is a different sort of variable -- a measure of output over a period of time. Throughput reflects the relationship between the capacity of the market and the capacity of the manufacturing facility.
As the chart above shows, the actual time that you spend running material through machines is often a small fraction of the total manufacturing time. For this reason CAD-CAM should not be considered a panacea for small manufacturing firms. Much refinement is usually needed before speeding up either machining operations or machine setup (the two greatest benefits of CAD-CAM) will provide the most substantial reduction in unnecessary production time.
When we look at the graphical representation of production developed above and analyze it in terms of activities, we find that most of the time was either spent moving material or setting up tools.
There are other important uses for this kind of information.
Consider the impact of the listed activities on your shop's resources.
Your organization has several kinds of resources available: labor,
machinery, inventory, material, and capital. This single operation
ties up quite a few of them. In fact, as the next graph shows,
this one operation monopolized a large amount of the total operation's
resources for a substantial period. Given more operations running
simultaneously, "collisions" (such as waiting for tools)
due to conflicting resource requirements are inevitable.
Eventually, in spite of your effort to improve the efficiency of your operation, you will maximize the amount of product you can generate with the available resources. At that point you will be faced with a difficult decision. You will have to increase your capacity. You will either have to expand your operation or subcontract some of your work.
WoodNet and the MTC offers a multi-faceted vision of what integrated manufacturing might mean on the Olympic Peninsula. As a flexible manufacturing network, WoodNet makes possible a much higher level of cooperation and communication than could exist in a traditional, competitive environment. It allows individual companies to team up and cooperate as though each was a division of a much larger company, pooling manufacturing capacity to tackle projects from outside the area and/or projects too big for any one of them to have handled alone. Other advantages include the possibility of directing the products of one operation into the inputs of others, or connecting one operation's waste-stream into another's source of supply.
If individual companies are to pool resources efficiently, the ability to adopt compatible manufacturing methods between operations is critical. In order to work together efficiently, the cooperators need to understand each other's operations very well and need to use communication tools that allow production information to be exchanged efficiently. This begins with developing a certain amount of common vocabulary, some of which is provided by this manual.
The MTC enhances the network by creating a wide-area-information network, tying the area's secondary wood products manufacturing businesses together through newsletters, meetings, classes, a computer network, and a centralized shared production facility.
The MTC would allow modern manufacturing methods to be tested for applicability at a far lower cost to the community as a whole than if everyone had to buy some or all of the production machinery, software, and other equipment just to find out if the move into CAD and CNC would substantially increase the marketability of their products.
The lack of standards, which is the case at all levels of management, presents formidable obstacles to cooperation. Computer networks based on telephone lines offer great promise for linking companies, and vendors are rapidly positioning themselves to provide this service. However, The problems are not solved simply by connecting computers. In a real way, that's when the problems begin.
Even if two companies are using the same CAD package to develop detail drawings, the odds that files can be merged easily is small. Each company will have its own symbol libraries, drawing layer names, line and color assignments, and revision tracking schemes. There are no standards that address file naming conventions or file management policies, either. Some companies will use libraries of standard details, copied into the drawing as x-refs so they will automatically update if a change to a standard detail is required, while others will have freestanding drawings with all the information contained in them.
The MTC will provide the clearing-house from which specific skills and standard practices are developed and moved out to the larger community of manufacturers.
In the corporate world, partnering between manufacturers and suppliers has led to "rating" and certification of suppliers. This is done because security is needed to assure quality and to reduce inventories. Your company is staking its place in the market on something made by someone else. Developing and adopting standards through the MTC to provide assurance that the timing and quality remain as expected is more efficient (and far more likely to work) than a maze of individual contracts.
Scheduling inventory and material requirements
As the examples (and the videotaping exercise suggested above) have shown, the problem of scheduling is not a trivial one. In order to work without a large inventory on site, you must be assured that you will not "run out" of parts or material in the middle of producing a batch of goods.
There are many ways to approach this. The simplest way is to work project by project, ordering the materials you expect to need in advance of beginning the project. The problem with this method is that it assumes a number of things that are not true: an infinite supply of material, an infinite supply of cash, and instant deliveries. None of these things can be counted on in the real world.
The unexpected lack of availability of some part or material can create a major diversion and waste of time (eg. tracking it through alternate sources of supply), and if it is not available it can cause you to miss a delivery or could result in costly redesign, threatening the overall project. Likewise, cash may not be available to purchase the entire inventory up front. Therefore, resource requirements must be planned and shipments must be scheduled.
It makes a lot more sense to let your suppliers know what you are doing and what you need, so they can be working on tracking everything down before you need it. This kind of relationship with suppliers is generally known as "partnering" and recognizes the interdependence of manufacturers and suppliers.
The question becomes one of predicting the size of inventories needed to provide "safety". The need to maintain inventories remains, if only to provide buffers. The two basic variations on inventory planning and material scheduling are known as "push" and "pull".
MRP is a "push" system based on projecting
the material and supply requirements that are required to keep
production running at the desired rate. Multiple independent
production processes are controlled externally, in an effort to
coordinate their requirements and to assure that any outputs from
one process that are required as inputs by another process are
available when needed.
Production and Inventory Control Society definition (APICS)
The Japanese model (JIT / kanban) is a "pull" system, based on the actual demand created by the production process. A "pull" system could not possibly work without partnering between manufacturers and suppliers. Using this system, an item is produced (or withdrawn from inventory) when demand for a replacement item at a workcenter downstream creates a vacuum which "pulls" another item into production. This implies a batch size = 1 production ideal (the direct opposite of the operational efficiency model based on production of a continuous stream of product). One of the most important manufacturing concepts introduced by the Japanese was the recognition that in manufacturing, everyone down-stream from you, everyone you provide a part or service to, is your customer.
Implicit in this recognition was the right of the customer to revise or clarify the specification in order to increase the quality of the finished product, or the ease with which satisfactory quality is achieved. The call for rework almost always happens down-stream from the place the error occurred, because the error or defect is discovered by the customer downstream.
Since you are the "supplier" of your downstream "customers", the partnership of JIT extends throughout the organization, as well as out into the world of suppliers. The stated objectives of Toyota's system are cost reduction and efficient use of capital. Efficient use is defined as an increased capital turnover ratio. These goals are achieved through the elimination of "unnecessaries" (wasted effort) and by maintaining a continuous flow of production.
Production is kept moving through JIT production and "self-stop automation". Any employee can stop the production line from any point along the line if a problem is encountered. The JIT is based on production methods characterized by small lot size, short setup times, workers who provide multiple functions, and tasks scaled to finish within one cycle-time of the operation.
The information system utilized to track Toyota's JIT production is Kanban, which means "card". Toyota uses a two-card system: one for conveyance (withdrawal for inventory) and one for production. Cards are exchanged for items.
When an item is added to a subassembly on the line, a card moves back toward the source of that part. The work-center which manufactures an item is only authorized to produce a part when a card is present.
The cards, which function as place-holders for the items that have been consumed, contain a surprisingly small amount of information:
You look at the plans and calculate the cost of materials based on what you think you paid last time, and then pad your estimate to make a little room for inflation.
Based on past experience, you estimate the time it will take.
You allocate your overhead based on the total number of hours in your estimate and the number of hours of machine time that ought to be involved, based on how much time the machines spend running.
You estimate the cost of preparing the estimate and the bid and then add a contingency.
Then you realize that you're way over industry average for the bid, and that you just spent over 4 hours working on this thing and still don't really know what it will cost. But you know you can't ask that much.
You ought to be able to do better than that. You've been collecting detailed production time information for years on your payroll time sheets if nowhere else. The key to more effective cost estimation and bid preparation is the development of a way to access this continuous stream of accurate cost and time information. The tools that provide this information are known as project management systems. They were developed to allow efficient scheduling of resources, including money.
Among the most baffling problems faced by a small manufacturer is the scheduling of activities and expenses. People tend to over-react to the appearance of a crisis, and remain tolerant of day to day irritants even though the day to day problems add up to a much greater problem than the crisis in the long run.
Definition of a project: Any undertaking with a defined starting point and defined objectives by which completion is identified (Project Management Institute: Wildman 1986 )
The key terms here are defined objectives, defined starting point and completion. A project is a series of related tasks that must be performed to achieve the goals defined for the project. In the discipline of Industrial Design, the project most often produces a process that produces a product.
Tasks in the project are related to one another because each must be completed in turn, before the larger goal, completion of the project, is achieved. Tasks are larger than steps, and each task contains many steps. The project usually takes a substantial period of time to complete, and is usually too large for one person to complete alone. For these reasons, project management is necessary to coordinate the efforts and performance of others to complete the project.
The function of project management is to schedule tasks and allocate the resources that are available within an organization to allow the project to be completed according to its specifications in a timely, efficient and profitable manner. The organization's resources include people, machinery, materials and capital. The project manager (probably you) must schedule activities to ensure that resources are available when needed and that all operations are undertaken in the correct sequence to avoid sequential error or over allocation of resources.
The fundamental understanding of project management is that, without a project schedule, it is virtually impossible to know where you are in the project, or what to do next, or whether the resources you will need are available. If you do not know what you must do next, you cannot manage.
In order to create a project schedule, one must plan. Planning is an operation that prevents sequential errors in the timing of actions or conflicts in the scheduling of resources. Few operations that qualify as projects can be accomplished without planning. The cost of developing and administering a plan is normally far less than the cost of time lost or of undoing actual errors that result from attempts to operate without a plan.
Planning involves the decomposition of the project into discrete tasks and the decomposition of the tasks into discrete steps. It also involves ordering the tasks to ensure that output of any task or step that provides input for another task or step is available by the time it is needed.
Planning requires information. The information most critical to the development of the plan is a thorough understanding of the goals / requirements / objectives of the project. Scheduling of resources is based on estimations of time requirements for each task (or each step within a task).
Development of a master project schedule (MPS) requires the following steps:
Many approaches to project management have been developed over the past 30 years. They all begin with time management, and they all rely on the creation of a Master Project Schedule similar to the one described above.
The outcome of a project is a product. Working backward from the characteristics of the product can provide the objectives of the project. Objectives must be clearly stated, understandable, and attainable.
One of the most important characteristics of a project is a well-defined conclusion, and one of the objectives of the scope is to define the characteristics of the project at its conclusion. Objectives that exceed the requirements defined in the scope at the completion of the project must be added through re-scoping.
Objectives must be:
Relationships (input / output dependencies) between tasks must be defined.
Estimates are better if based on previous experience with similar tasks and resources.
Data used in estimation must be revised (re-evaluated / calibrated) at the close of project.
Estimates must be based on calendar time.
Base cost estimation on:
Create an independent scheduling timeline for each resource, including human resources.
Develop a schedule:
List all existing (prior) commitments for each resource. Include human resources.
Compare requirements of Master Project Schedule with prior resource commitments related to other projects.
Redefine conflicting tasks into smaller steps so that one of them can be begun without access to the resource.
Revise workflow to eliminate conflict.
This means formally reserve the resource
This requires a mechanism in place for scheduling resources.
Reconcile conflicts between task completion times and project timeline
Most efforts to coordinate manufacturing processes have been based on traditional project management models. Program Evaluation and Review Method (PERT) and Critical Path Method (CPM) were both developed to coordinate large projects. Each offers advantages and disadvantages for the small manufacturer, which will be addressed in a later section.
Both methods are based on the development of precedence networks. Precedence networks define the relationships between tasks. The idea of precedence is to determine which tasks must precede others, and which tasks cannot begin until the output of another task is received. This is the basis of scheduling, establishment of inventory sizes, and the ordering philosophy embodied in JIT scheduling. Both systems allow managers to identify resource conflicts and to experiment with resource leveling. PERT allows the manager to address a range of probable completion times, based on uncertainties in resource availability and performance. These methods will be addressed again in the next section.
The systems above require that you input a lot of information that you are not likely to have at your fingertips. Long before you can implement a project management system, you have to start collecting business information in a form that can be translated easily into your project management system.
Once you recognize that you need to know how long it takes and what it really costs you to do things, it's amazing how much information you can find.
The first problem is figuring out how and where to store it. Obviously, you need to be able to get access to it quickly and easily, without having to tear through cardboard boxes of old papers or even old computer disks. And yet, you can't afford to spend money collecting data you may never need.
The solution to this problem is to create a structure to hold information that can serve multiple purposes.
The long term solution is to implement project management software as your organization's communication backbone. Unfortunately, the resources to do this are not likely to be available. The short term solution is to create a structure that can be used to feed information to other solutions. The solution I propose is called a Product Data Manager (PDM), a tool which comes from the MRP / MRPII model. The reason it is a useful place to start is that it provides an easy to understand structure based on products, broken down into parts. The parts are broken down into tasks (the steps involved in making and assembling them).
To assemble this tool you will probably want to start by outliniganize it automatically
Large companies establish entity-wide databases which collect and access business information and tie time management, accounting, and other systems to this database. Access to the data is provided by query and the emerging standard means of query is Structured Query Language (SQL). SQL provides the nervous system of your business. Most small businesses begin and end their records management systems on tracking money rather than tracking time.
As of this writing, there is no easy-to-implement, "off the shelf" solution to integrating time and resource management and money management for a small operation.
This does not mean you should abandon the idea. Need spawns solutions in the software world. The need is probably already nearly filled. By the time you have the collection of data underway, an easy way to organize and integrate it will have arrived.
In spite of the fact that you will eventually need to develop a databse, I believe that a spreadsheet program is the easiest tool to use to start collecting management information. They currently provide the easiest means to move data from one format to another. All the data you collect will be transportable to the database, once you figure out how you really need to organize it.
The issue is to start collecting data first, then, once you are using it, to figure out what you really need to do with it. Speadsheets are the easiest tools to use to toss clumps of information around. Until you start doing real analysis, the overhead of a spreadsheet is minimal and the line between the capabilities of spreadsheets and databases continues to shrink (as spreadsheets grow more and more data handling capabilities). All the major spreadsheets can read data from external databases.
The integration of stand-alone programs into "suites" of software is an ongoing lesson in frustration. The real, enterprise-wide data management solution you need for a small firm does not really exist, but awkward solutions are out there: (the original model was DDE (dynamic data exchange), which has been superceded by Object Linking and Embedding (OLE and OLE2)). Through OLE, programs like Microsoft Project® allow Microsoft Excel® spreadsheets to be attached as "objects" within tasks. Objects are an integration of data and structure. Object linking provides these objects with live links back to the program that created them. If the file that provided the source of the information in the object changes, then the object will change to reflect these changes the next time the application containing the embedded object is opened.
OLE allows links to be maintained between your Project Manager and all the