The ECCO System

Cybernetic Principles for Effective Control in Complex Organizations

Click here for footnotes, Ch.7


Conclusions, Significance and Directions for Future Research


Evidence for the Principles

Each of the companies investigated showed striking examples of the use of all of the cybernetic principles which have been identified as ECCO System practices. The following is a summary of the specific evidence for each principle found in each of the Baldrige Award companies assessed.

Closed, shortened feedback loops (Principle 1)

Motorola Corporation has shown itself to be explicitly aware that "open loop" processes cannot achieve more than three or four sigma performance. The entire strategy for quality improvement at Motorola is the routine, continuous tracking of all information which is deemed important for quality. The day-to-day tracking of this data is performed by the workers, who are empowered to analyze and improve both process and outcome. Motorola has also aggressively incorporated its suppliers into product development efforts, routinely including them in portions of the planning and development process. Moreover, Motorola has undertaken intimate involvement with supplier organizations' internal quality efforts, requiring high quality standards and processes and providing education, training and assistance to assist suppliers in meeting those demands. In addition to formally tracking customer satisfaction, Motorola management and service personnel maintain personal contact with operational level workers in their customer organization to discover, track and improve both products and business processes such as order delivery and invoicing. Customer input is routinely sought by Motorola teams involved in the development of new products, and customer feedback is solicited periodically during the development process to ensure the development efforts remain on target for customer needs.

IBM - Rochester has hundreds of measurements at all levels to track processes and results considered important to the corporate mission. Responsibility for identifying, tracking, and improving these processes rests with individuals or groups of workers engaged in the local process. In general, responsibility has shifted downward, with more responsibility and authority given to operational personnel, as measured by the employees' responses on an anonymous, annual questionnaire. All line employees are empowered to shut down not only assembly lines, but the entire plant if necessary, to correct quality problems and have, in fact, done both at least once in the last two years. Customer feedback loops are closed as well. In product development, customer input is routinely solicited from standing Customer Advisory Councils, and customers are brought in by new product development teams to test simulations of new products during the development process. Rochester operational employees follow up within 30 days of purchase with every customer who buys an AS400 system, for feedback on customer satisfaction, and to resolve problems which may have occurred. Perhaps the most impressive change of IBM policy at Rochester has been the modification of IBM's legendary proprietary confidentiality. Now, new product development teams share plans and consult with suppliers, beginning early in the development process, using suppliers' expertise and feedback to modify designs.


At Xerox Products and Services, "closed-loop" practices are formally integrated into everyday efforts of Xerox personnel. Routine tracking of both corporate-identified and locally-identified processes essential to quality is done by operational personnel and first-line managers. These people have the responsibility for analyzing and developing process improvements, and have the authority to implement those improvements in co-ordination with local "process owner" managers. Sales personnel have increased authority to negotiate price directly with customers. Site service personnel may grant waivers of up to $300 worth of repair charges on their own authority, or waive charges entirely if they believe the problem was due to prior errors on the part of the service department. District offices have the authority to directly approve customer credit, and to grant up to $10,000 customer accomodations without higher approval. Customer satisfaction with Xerox products, service and administration is routinely monitored by monthly surveys and results fed back to district and local offices. Xerox actively involves customers and suppliers formally and early in the Product Development Process directly with development teams, and obtains customer evaluations periodically throughout the development process.

Improvement of capabilities of first-line regulators (Principle 2)

Motorola Corporation has undertaken to provide extensive education to their lower level personnel to bring basic mathematics and communications skills up to required levels, and provides specific formal training for all employees in new methods for quality improvement. For design engineers, this includes Design for Manufacture and Assembly. For software engineers, it includes structured methods for producing "reusable" code. Everyone is trained uniformly in basic statistical analysis; the quality improvement process, principles, techniques and objectives; and in group process techniques, in addition to receiving specific, job-related training. Along with the training and education comes empowerment: Every employee has the authority and responsibility to identify areas in need of improvement, and are routinely authorized by first line managers to form teams to investigate and remedy processes in need of improvement.

At IBM - Rochester, non-management employees are given specific job-related training on the job, and in both in-house and external classroom settings. Additional in-house courses train manufacturing employees in statistical techniques and group process, and provide cross-training appropriate to the employee's work environment. "Cascade" training has now been formally adopted by IBM (based on the success achieved less formally at Rochester) which covers terminology, concepts, vision and philosophy of the IBM Quality Initiative, as well as process analysis and management. Line and operational support employees are expected to recognize, analyze and (in co-operation with local management "process owners" when appropriate) correct processes contributing to quality problems.

Xerox Products and Services has an explicit focus on training in three major areas: Job functions, responsibilities and special skills associated with these for all operational personnel; "Cascade" training for all employees which includes an introduction to the Leadership Through Quality program, standardized basic statistical techniques, the Nine-Step Quality Improvement Process, the Six-Step Problem Solving Process, benchmarking, and Cost of Quality calculations; and finally, team processes and group techniques. The latter are modeled in the "cascade" training, with additional training available in-house. This is often provided as special skills training, especially for standing work groups and product development teams. Work groups have vastly increased authority over their internal management and problem-solving, and operational employees are trained and empowered to correct processes in need of improvement.

Heterarchy of first-line regulators (Principle 3)

Motorola Corporation considers teamwork to be the backbone of their "Six Sigma" quality efforts. For the development of new designs, a multifunctional team of managers deteermines what functional areas must be involved in the effort; then appropriate staffing and budget are provided. The identified design development team itself consists of operational personnel from all functional areas which touch the process, from design engineering through sales and shipping. Their design plan is reviewed by an independently staffed cross-functional review team, composed of operational personnel and front-line managers. Cross-functional problem-solving teams are constituted of operational level employees from every functional area which touches the process to be improved, who are responsible for all aspects of the solution and are empowered and responsible for its implementation.

At IBM - Rochester, the new product development teams involve representatives of all departments which might impact or be impacted by the development and shipping of the new product. Traditionally an informal process, team membership and representation criteria are now being incorporated into a more formal protocol. Process improvement teams at IBM - Rochester typically have open and shifting membership, and tend to includee a mix of line employees, technical personnel and managers. These teams may contain more than the representation required by the heterarchy of first-line regulators, and might conceivably contain less than full representation. However, when they identify a problem and potential solutions, they work with the designated "process owner" and personnel involved in all aspects of the process to implement and test the proposed solution.

Xerox Products and Services product development teams are designed to bring operational personnel and resources (including suppliers) together early in the planning and development phases. Consultation and approval of financial requirements is incorporated into the development process, but no longer drives the process as it did in the 1970s and early 1980s. A highly formalized process, the Product Development Process is designed to ensure that designs and their impact are fully modeled in each phase by operational personnel and their first line management, and problems ironed out before proceeding to the next phase of development. Problem-solving teams are also used in both manufacturing and business processes. "Process owners", managers accountable for process improvement, bring together representatives from all departments involved in the process to analyze and implement improvements.

Analysis and correction of process (Principle 4)

As previously noted, Motorola Corporation's entire focus is in analyzing and correcting processes. They use the technique of benchmarking both for assessing the level of performance to aim for and to determine the processes used to achieve that level of performance. Extensive process analysis in all operations, from Just-In-Time inventory and production techniques through the paperwork processing of orders and inventory. They emphasize the primary importance of process improvement to achieve the "entitlement" level of corporate performance prior to investing in new information and automation technologies. The first job of problem-solving teams is to analyze the process under investigation by formally mapping it. They then proceed to identify bottlenecks and non-value added steps, which they are empowered to eliminate by redesigning or modifying the process.

Process correction is likewise the standard at IBM - Rochester. The focus explicitly shifted from product (doing whatever would ensure the quality of the product on an ad hoc basis) to process (analyzing, refining a reliable process to obtain excellence of outcome) in 1989, as a result of the Baldrige Award Committee feedback.

It is clearly acknowledged at Xerox that to improve results, you must improve the process. Approximately 75% of all benchmarking efforts focus on processes rather than the service or product they produce, and they emphasize that a valuable "lesson learned" is that benchmarking should be used primarily for process improvement rather than setting goals or quotas. In addition, the entire problem-solving approach which is standard at Xerox consists of process analysis and correction, from the initial flowcharting to the final assessment of success.

Role of management (Principle 5)

Motorola Corporation provides extensive management education to develop a pro-active management style in support of quality culture. This approach involves management understanding of how to support and coach the workforce in the use of tools and group techniques for problem-solving, and how to co-ordinate and co-operate with other managers. Top management is actively fostering not only education, but evaluation and rewards in support of employee empowerment, and the creation of a common environment in the corporation with specific policies to ensure that everyone, at every level, acts to support Total Customer Satisfaction.

IBM - Rochester is working on changing managers from being "detail-specifiers and doers" to being goal-setters and coaches. Specifically this involves management giving responsibility and authority to subordinates while retaining management accountability. The change in management style, supported by new understandings about methods for the achievement of quality seem to have fostered the environment for increased employee participation and aligned incentives.

Xerox has had to overcome its culture of the 1970s, which was characterized by the precedence of financial management by the numbers, and a somewhat antagonistic relationship among management and labor, and between departments. In contrast, the Leadership Through Quality initiative has recharacterized the role of management as one of leadership, setting goals and providing context and guidance, and fostering of co-operation among departments (and with labor unions) to achieve overall corporate goals. It has consistently moved responsibility and authority downward into the hands of trained and empowered employees, eliminating two layers of management in the process.

Alignment: Incentives, evaluations and rewards (Principle 5a)

At Motorola Corporation, all employee evaluation policies are being reviewed and modified to support the goals of "Six Sigma". The evaluation of managers is directly tied to their participation in crossfunctional quality efforts and outcomes rather than being tied to the success of their individual department, as in the past. They are also evaluated on such internal support of the "Six Sigma" effort as their encouragement of participative management in their departments, and fulfilling their employee training commitments. Senior executives are now evaluated on the success of their co-operative efforts to achieve corporate goals, not success in their individual functional areas. Hidden incentives, such as those which encouraged salespeople not to take the time to process new orders until the end of the week, are being discovered and altered to support rapid processing, the desired, customer-focused outcome. In short, everyone is evaluated based on their contributions to the achievement of the main corporate objective: Total Customer Satisfaction. Financial success is shared via profit-sharing, and special monetary awards for outstanding individual and group contributions to quality. In addition, Motorola makes special efforts to provide recognition for individuals and teams by spreading the word in formal and informal communications media. And outstanding contributions to quality and fostering successful teamwork are often rewarded by promotion to management positions.

As a part of their corporate Market-Driven Quality Initiative, IBM has identified a number of key processes targeted for improvement at the corporate and subsystem levels. To ensure management commitment, each process requires an executive level or management level "process owner". If the expected improvement does not occur, the "process owner" loses a portion of his compensation. Since improvement of most processes requires interfunctional co-operation and active participation of the operational personnel, it is a powerful incentive for the management process owner to support and encourage these activities. IBM's evaluation system for all employees below the top level of management (Performance Planning, Counseling, and Evaluation or PPCE) involves individual negotiation between the employee and his supervisor. It appears that as MDQI values are incorporated at Rochester, PPCEs are converging to reflect and embody those values, so the alignment is taking place without being formally and uniformly mandated. In addition, a culturally powerful mechanism exists which influences the behavior of managers: feedback on their performance by their subordinates in the form of the annual, anonymous survey completed by all employees. The combination of this feedback, employee training, commitment to process analysis and improvement, and "process ownership" accountability is effectively aligning the incentives of the organization.

Xerox has undertaken extensive formal changes starting at the top of the organization to align incentives and evaluations. Xerox executive compensation criteria have shifted away from rewarding functional expertise and accomplishment, and instead rewards all executives based on overall company success and customer satisfaction. Line managers are evaluated based on successful implementation of quality processes and on effective human interactions with subordinates as well as cooperation with other managers. Where a potential conflict of interest existed among sales, service and administration, these departments were brought together in partnerships from the local level to that of Vice President, where evaluations for raises, bonuses and promotions are heavily based on the achievement of their common goal: customer satisfaction. This has encouraged co-operation, rather than competition. Formal recognition of excellence in teamwork is provided by way of Team Excellence Awards, many of which carry a cash bonus. In addition to spreading success stories in corporate communications, a powerful form of recognition of teamwork is provided by national and regional "Teamwork Days", exhibit conferences where teams can display and share their successes with other Xerox personnel as well as the general public.

Alignment: Measurement systems (Principle 5b)

Motorola Corporation takes as axiomatic that you must measure everything which is important to quality, reliability and economy. They developed their definition of "Six Sigma" quality (approximately 3.4 defects per million), and use this standard for measurement of all defects and errors. They also measure cycle times for all processes, both in manufacturing and in support areas. The Six Sigma Institute has a group specifically devoted to examining and reformulating and validating corporate measurements and models. One of their recent accomplishments was the reformulation of the cost accounting model. They have also systematically changed measurements throughout the company from focusing on internal criteria to measuring what is important to the customer. For example, instead of measuring the difference between the time a shipment was promised and the time it actually shipped, they now measure the length of time between the customer's placement of the order and the time it is received.

IBM - Rochester officials acknowledge that "the organization you have is a direct result of the measurements and evaluations you use." A specific set of measurement and evaluation criteria havee been incorporated in writing for design and manufacturing processes. Although no formal requirements for specific measurements or analyses exist outside of the manufacturing area, a variety of data-based analytical techniques are used in support areas such as Administrative Services, to analyze and allow improvement to key processes. These areas are held to the same order of magnitude improvement as design and manufacturing. The emphasis on key processes improvement guides the development of the measurements.

Xerox's new Business Plan radically changed what was to be measured on a day-to-day basis. These included key external measurements such as customer satisfaction with products, services and administration, as well as internal operations measurements of quality and cycle time, all of which were designed to be congruent with customer requirements. The new measurement system replaced standard quality (uniformity) and productivity (man/hour) measurements dominant at the company into the early 1980s. Perhaps the most critical change was the institution of the Cost of Quality measurement, which takes into account not only how much it would cost to do something right, but also the cost in customer satisfaction of doing it wrong, and the cost in lost opportunities of not doing it at all.

Alignment: Company goals (Principle 5c)

Motorola Corporation provides extensive training, education and continuous communications at both operational and management levels about its company goals, processes, and values, termed respectively key goals, key initiatives and key beliefs. In addition to uniform training and follow-up with company personnel, understanding and practice of these is supported by measurements and evaluations which reflect corporate goals. Also critically important are the periodic audits conducted by top management teams which are designed to ensure that divisions and departments are not only measuring and achieving the goals set, but doing so in accordance with key beliefs and key initiatives (processes).

At IBM - Rochester, company goals are shared formally and informally, based on a strong and pervasive internal culture, and supported by the official local and corporate communications in electronic and journal form. On-line systems contain company information and measurements accessible to all departments, as well as a common electronic communications system which reinforce informally the company goals, objectives and means of obtaining them. Alignment at IBM - Rochester is highly intrinsic, supported by excellent morale, a flexible set of working incentives and evaluations and a people-oriented management attitude.

At Xerox Products and Services, company goals are not only shared in official publications and uniform management rhetoric but are deployed through the organization by the Business Plan measurement system, the uniform application of quality tools and problem-solving processes. They are reinforced by the management "process owner" system and the employee evaluations system which ties rewards and incentives to the achievement of corporate goals. Paul Fichbach claims that one element which impressed the Baldrige Award examiners was that every employee could say how what they were working on supported which company objectives, from the local level through the presidents-level goals.

Alignment: Common language (Principle 5d)

Motorola Corporation is a clear example of a company having a consciously adopted common language for their operations. Motorola's "Six Sigma" measurement theory and practices provide standardized metrics and terminology for all corporate operations. Their calculation of Parts Per Million (Total defects per operation divided by Opportunities for defects) translates the definition of Six Sigma accuracy for use in support and administrative areas. In addition, the Key Corporate Goals, Key Beliefs, and Key Initiatives, along with common training and uniform tracking mechanisms by Quality Audit provide a concrete, commonly understood and utilized vocabulary for the entire organization.

While IBM - Rochester has adopted some special and formal terminology associated with its quality efforts, it is somewhat an exception in terms of having a specific, detailed language and system of common metrics peculiar to their quality efforts. The plant seems to rely on a strong and embedded culture rather than on explicit uniformity of language and concept in its quality efforts, although language associated with IBM's Market-Driven Quality Initiative is being incorporated in formal and informal communications.

Xerox has developed not only standard terminology in its pursuit of Leadership Through Quality, but common methods with their own technical language (Quality Improvement Process and Problem Solving Process), and the common use of benchmarking and Cost of Quality analysis. These vehicles provide a common way of viewing and communicating data throughout the company and across department and division barriers.

Summary of Expected and Unexpected Findings

In spite of differences in corporate history, market position, internal culture, and employment conditions, evidence for all of the ECCO system principles was present to an impressive degree at all three companies visited. IBM was the only company which displayed any significant variance, and this was in the area of explicit incentives and evaluations. This difference was explained by IBM managers as the presence of the effective incentives embedded in IBM culture, which had the same effect on employee behavior as the more explicit policies undertaken by Xerox and Motorola. Similarly, all the companies have either eliminated layers of bureaucratic and supervisory hierarchy or at least decreased the number of first line and middle managers, increasing spans of control.

The research, somewhat surprisingly, did not present any serious challenges nor potential refutations to aspects of the ECCO system theory. It did, however, highlight two new aspects of potential importance. The first is the power of uniform training of all employees in basic concepts, skills, measurement protocols, and vocabulary. According to company representatives, this uniformity enabled the communications among departments and across levels to be enormously more clear and effective. By creating a lingua franca for both communication and measurement criteria, a whole host of misunderstandings was eliminated among different functional areas, along with problems of how to compare efforts and successes in different departments. This also provided a new, shared conceptual framework essential for the support of the new unified system focus which is the basis of Total Quality efforts.

The second aspect discovered in the research, which may deserve to be a principle in its own right, is that of the nature of oversight inspections. This may be critical only in the transformation phase; however, it is well worthy of further research. The monitoring and inspection which is done by the top level management of all company operations is monitoring of appropriate meta-processes--the way decisions are made and planning and design are done--and the specific results obtained by carrying out the plan or implementing the design. This is in contrast to most inspection characteristic of Weber/Taylor bureaucracies, which either looks at the end results alone, or at the process itself for conformance to approved norms. And sometimes it looks at both: it requires a process to be performed a particular way and also requires certain outputs as a result, a fundamental error. The problem with specifying both process and outcome simultaneously is that when the input changes or the environment changes, most processes will produce different output from the original, desired output. Corroborating this observation, Xerox PARC researchers have found in formal empirical studies that most office employees studied used highly original methods for dealing with a wide variety of document processing problems, rarely following officially mandated procedures because those procedures didn't work for most situations. They referred to the formal procedures for criteria to help them develop adequate solutions for the problems at hand. Anecdotal evidence for the inefficiency of standard procedures in bureaucratic organization also abounds.

In the companies studied, the meta-processes were determined and taught to operational employees: problem identification and solution development protocols, statistical process control methodologies, specific process analysis methods. And improvement goals for output were set, usually 10 times and 100 times improvement within a specified period of time. The top level quality management team would review progress in identified operational departments or interfunctional areas every 6 months to 1 year. These reviews (as well as informal "check-ins") were designed to assess progress toward the goal, and to be sure that the problem analysis and solution steps and methods were being followed. The operational processes and practices developed as a result were quite flexible. The monitoring of the meta-processes seems to ensure that the cultural and infra-structural standards which create the environment for ongoing interactions in the company are maintained. This approach has both the advantage of leverage of authority, eliminating the need to micro-manage, and perhaps more important, it eliminates the problems associated with specifying both process and outcome simultaneously.

In terms of cybernetic theory, this principle can be traced back to Ashby's work on the homeostat: The adaptability of the homeostat to a wide variety of sets of initial conditions rests on the fact that it has a stable meta-structure, and a variable internal structure, and a required set of output conditions. This principle of stable metastructure with variable internal structure is currently found in the control of other complex systems as well, most notably automated intelligent support AI systems, and so may well be of the most general importance and applicability.

Revised Synopsis of Applied Principles: Foundations for ECCO System Infrastructure

Although it is widely recognized that bureaucratic organizational structures are an impediment to Total Quality efforts, little explicit attention has been given to the way control is actually achieved in "transformed" Total quality organizations. Most organizations in U.S. industry and the military are Weber/Taylor bureaucratic hierarchies. There are several structural and operational features of this type of hierarchy which directly oppose some of the features necessary for successful implementation of modern technologies and practices associated with Total Quality efforts.

It has been the suggestion of this work that the core of this problem is that the Weber/Taylor system was constructed for dealing with large complicated linear problems, and is structurally incapable of dealing with COMPLEX, non-linear problems which characterize current challenges to modern manufacturing organizations. A critical question is "What control structure will adequately replace the Weber/Taylor structure?".

What has been claimed by the current thesis is that the success of TQM lies in the utilization of specific measurement and analysis techniques, and enabling technology functioning within a particular co-ordinated infrastructure. Without this infrastructure, most of the effectiveness of the techniques and the technologies is lost. Such an infrastructure is referred to as an ECCO system structure (for Effective Control of Complex Organizations). The nine principles here do not themselves fully constitute the infrastructure, but they form a basis for its construction. Hierarchy in ECCO system (or TQM) organizations becomes flattened, since it is no longer used primarily for monitoring and controlling internal activities, and begins to resemble "natural hierarchy", which is based on time-scale of its decision-making rather than span of control.

Principles of the ECCO System

Nine major principles have been identified which are of primary importance in the implementation of Total Quality efforts. The first four apply to the operations of the organization (regulation) and following five to the management of the organization (control). They can be stated as follows:


1. Close and shorten all feedback loops. There is no possibility of regulation of a variable if feedback loops are not closed. The effectiveness of regulation and adaptation depends on timely feedback and action. This means that the authority to act is always located in the lowest possible echelons.

In application, this means that everything important to the achievement of company goals must be measured routinely, and that people involved in the process should both measure and be able to correct problems and improve performance. It also implies that customers and suppliers, who are functionally a part of the company, should be involved firsthand in product development; and that customer satisfaction must be routinely tracked, analyzed and have problems remedied by people as close to the source as possible, since corporate survival depends on it.


2. Maximize the capability of first line regulators. This entails a) providing the necessary training specifically related to job requirements, monitoring and analysis techniques and b) the direct authority to change (regulate) whatever it takes to achieve the local goals.

In application, this means that all employees with operational responsibilities must be trained and empowered to act to correct quality problems they detect.


3. Planning and problem solving is done by a heterarchy of first line regulators who (in effect) model the process under revision or development. (This usually entails inter-departmental or crossfunctional participation).

In application, this means that planning, including product design, and problem-solving can be done most efficiently by a group of representatives of each of the operations which affect or are affected by the process. The team members consult and negotiate among themselves, knowing (and sometimes modifying) their departmental requirements, impacts, and resources, to develop a final product or solution which is at least fully acceptable to all.


4. Analyze and correct process, not just inputs. The outcome of a complex process (one with circular causal chains, non-linearity, etc.) can be equifinal, obtaining the same results from radically different inputs; or may give radically different results at different times, given the "same" input. Until you have analyzed the process, you cannot control the output.

In application, this means that to solve a problem or improve an output, the process by which it is currently achieved must be mapped or flowcharted, analyzed, and modified until the desired output is achieved. This is in sharp contrast to the practice of finding an individual to blame, or exhorting people to work harder or faster.


5. Intrinsic alignment of incentives, evaluation system and reward structure to encourage the behaviors which will achieve the overall goals of the system as identified strategically. This may mean a radical re-evaluation of the dynamics of co-operation which is required to solve problems which occur across departmental boundaries, and across levels of hierarchy.

In application, this means that employees must be evaluated and rewarded for contributions and co-operation which helps achieve the overall company goals. In addition, management must discover and change any conditions which make employees want to do things which interfere with company goals. (These are usually not apparent at first glance.)


6. Intrinsic alignment of all measurement systems, so that what is being measured (controlled) is supportive of or important to the achievement of the overall system goal.

In application, this means that the measurements being tracked must support the achievement of, for example, quality or customer satisfaction. This is in contrast to the practices of measuring such things as productivity in terms of parts per person-hour, and rather measuring (and minimizing) time from initial order to delivery.


7. Intrinsic alignment of company/system goals throughout the organization. This means having a set of mutually congruent strategic system-level goals, and ensuring that all sub-goals of departments, and of interdepartmental co-operation projects are congruent both with the system goals and with each other.

In application, this means that first, the overall goals of the organization are well understood by all members of the organization. Second, instead of each department having their own goals to optimize their department function (usually at the expense of other departments' performance), department goals are based on their co-operative contribution to the overall goals of the organization.


8. Intrinsic alignment of communication system by introduction of a common language and general evaluation protocols.

In application, this means that common terminology relating to the achievement of quality and customer satisfaction is shared by all members of the organization, as well as problem-solving strategies and related tools, and uniform metrics or methods of measuring progress and goal achievement.


9. Oversight inspections to ensure that common meta-structure and meta-processes are used in all operations and that goals are being met, while processes used to obtain goals are allowed to vary.

In application, this means that top level management audits departments within the organization, to be sure they are using the common methods to solve problems or perform planning and design activities, and common metrics and appropriate measurements to evaluate progress, as well as achieving their goals.

Thus, the ECCO system model distinguishes between regulatory activities, which are operational, and control activities, which are co-ordinative and strategic. Control is a function of the creation of an environment which intrinsically supports behaviors favorable to goal achievement. Effectiveness and efficiency are a function of the regulatory structures which maximize the effective variety of regulators as close to the source or element to be controlled as possible.

The bureaucratic model consists of an environment and structure which are effectively aligned in opposition to desired behaviors, such as co-operation. The traditional methods of bureaucratic control (inspections, documentation, span of control and vertical reporting structures) when applied to modern complex organizations only make performance worse.

One reason why top management commitment is so critical to success in Total Quality efforts is that only top management can change and create new environments within the organization. If top management gives verbal support to TQM ideals, but requires the carrying out of techniques and implementation of technologies without this fundamental restructuring, they are truly failing in their leadership commitment.

Boundaries and Limitations

Before discussing the significance of this research, it is important to point out what it does NOT purport to do. This will include a number of issues which may be vitally important to Total Quality efforts. Guidance on these issues must be sought from independent sources. The following is an outline of some, but not all, of the issues explicitly not addressed (or not addressed directly) by the present work.

The ECCO system model is essentially a control system model, which has implications for a number of specific internal practices and structures. It leaves out the critical issue of how to create the goals and visions which guide the company. Elucidation of this issue properly rests with some strategic management literature in terms of matching the organizations strengths and capacities with the content of strategic goals. The process by which these goals and visions can be developed also falls into the category of "self-organization". Heinz von Foerster and others in second order cybernetics have some major contributions in this area, as do Russell Ackoff, John Warfield and associates and Peter Checkland among others.

The ECCO system principles are neutral with respect to the specific set of analytic tools and processes to be applied in measuring and improving quality. Statistically based tools have been shown to be invaluable in measurement and improvement of physical systems and have been shown to be applicable to a variety of semantically based systems (such as billing statements, interpersonal interactions such as sales etc.). However, the ECCO system criteria remain pragmatic: The measures should fit the situation and the variables considered to be important. Simply because it is possible to apply statistical analysis to a problem does not mean it should be applied, nor that it will be appropriate vis-a-vis the problem under consideration.

A major aspect which is left open, but is critical to corporate survival is that of company evolution based on corporate learning. Several researchers in System Dynamics have addressed this problems, most notably Peter Senge in his new book about the importance and nature of organizational learning. System dynamics modeling, and associated systems thinking methods seem to be critical to examination of complex causal processes, as well as to learning at the organizational level as a whole.

The role of lower level employee input to strategic level goals and control is also left open. The ECCO system principles are neutral with respect to this: Japanese companies rarely solicit input from their lower level employees as to company-wide policies; American companies tend to aim for more input, and share some critical corporate decisions with hourly employee representatives as in the case of Xerox joint union-management analysis of plans to contract out work previously performed in-house. One company based in South America is so democratic, they involve all level of employees in budget and salary decisions, and allowed manufacturing employees to overrule top management on the selection of a new plant site. All of these companies, however, may meet ECCO system criteria. Similarly, Quality of Work Life issues are not addressed explicitly by the present work. The specific conditions of employment and quality of life policies will be influenced by the relationship of employee expectations with the policies which will fulfill them conjoined with system goal achievement: What policies will fulfill the conditions of alignment in incentives, rewards and evaluation.

Closely associated with this is the more general consideration of the adequacy and structure of communication channels among various levels of employees. All of the companies studied had direct line access from any employees (who could remain anonymous if they wished) to the CEO (who was required to respond). This may or may not be critically important. It certainly had a positive effect, and allowed important issues to be raised and resolved rapidly, enhancing customer service and employee satisfaction.

The ECCO system principles deal only peripherally with the adequacy of communication channels. One systems modeling approach which deals explicitly with the issue of communications channels (and which contains some overlap with ECCO system principles) is Stafford Beer's Viable Systems Model (VSM). The VSM, also based on work by W. Ross Ashby, views the organization as a non-nested recursive hierarchy of five systems, each performing one of the operational, co-ordination or planning activities necessary for ongoing survival (viability) of the organization as a functioning entity. The VSM can be used to analyze and correct various aspects of an organization's functioning, and deals particularly well with both ensuring the functional presence and the co-ordination of (communication among) the five required activities.

As noted previously, the ECCO system principles do not indicate levels of expertise, resources or reinvestment requirements necessary for corporate survival. One serious concern in the literature, for instance, is the quality of engineering design: Specifically, the concern is that the United States industrial community does not have the educational resources or expertise to train new engineers in the content aspect of quality design. These, and many other issues like them, are beyond the scope of this work.

Finally, the ECCO system framework has not specified how or where the cost savings are achieved. Certainly savings are achieved by eliminating errors and clearing bottlenecks from processes. Some major financial benefits accrue to companies as a result of increasing market share and adopting technologies which increase their capacity to make money. It has been shown to be critical that non-value added steps be minimized in operations. However, this apparent limitation in the theory may be appropriate as the principles may be applied to non-manufacturing systems. This may include not only service organizations, but those which operate on voluntary contributions; on fixed or shrinking budgets with expanding mandates and requirements; and those which have multiple customers with non-aligned or potentially antagonistic interests such as medical patients and third party payors. To some extent, different sources of savings, and different criteria for quality, are likely to apply to such organizations.


Having clarified some of the explicit limitations of the research, it is now appropriate to consider the significance and potential contributions. This work has identified several aspects of bureaucratic practice which impede Total Quality efforts, and has proposed the mechanics of a replacement control structure. This means that organizations can analyze their operations and correct deficits and errors in their TQM implementation.

This theory proposes a radical redefinition of Total Quality Management. The specific techniques and technologies associated with modern manufacturing and quality improvement are critical adjuncts to or direct instantiations of the ECCO system principles. The ECCO system infrastructure provides the framework within which the techniques and technologies can operate effectively. No other author thus far has provided a theoretical infrastructure for TQM which specifies the control mechanics for management, along with criteria for adequacy of techniques and practices which should be required, and why.

The ECCO system approach redefines TQM in terms of a system adequate for the achievement of specific goals under changed environmental conditions: limited time, limited resources, expanding and volatile output requirements. The associated technology systems are 1) a reflection of the principles; 2) enablers of improved performance; and/or 3) in the case of integrated technologies, are systemically designed to solve current problems, and therefore require an ECCO system reconfiguration of internal structure in order to operate as designed.

Now it has been shown not only what traditional management and control functions must be given up, but why, and what ought to replace them. Thus, this work has developed the skeleton for a new paradigm for integrated control in organizations.

TQM disappointments and struggles far outweigh their unqualified successes. This is particularly true in the auto industry and the Department of Defense, where Total Quality programs are mandated for all contractors, Federal Contract Research Centers, and the military. To the extent that this theory is valid, it can provide assistance in diagnosing and remediating systemic problems in the approach.

Directions for Future Research

There are several research paths which it would be both interesting and fruitful to pursue. First, the theory deserves a wider test of its principles. To what extent do these principles apply to other successful Total Quality organizations? Are the principles themselves a good indicator of Total Quality, or are they a partial answer only? To what extent do the principles apply in the for-profit service sector? Do they apply to organizations with third party payment systems, government regulated profits, or organizations which have multiple customers? Another question will be to what extent can organizations such as the military adopt this non-bureaucratic form of control? What are the limitations on organizations who adopt part, but not all, of the principles? In addition to more extensive testing of the principles and their application to a wider variety of organizations, it will be important to develop and test methodology and specific instruments to diagnose ECCO system deficits in organizations.

An area which could clearly benefit from more detailed examination is the configuration and leadership of the various groups which instantiate the regulatory heterarchies. Some group, especially permanent work groups, are leaderless. Some are highly participatory in an apparently egalitarian way, but must convince or satisfy a specified "process owner" of their recommendations. There is some evidence to suggest that expert senior technical leadership of design groups (concurrent engineering and project task force groups) produces superior results compared to other forms of organization leadership. It will be fruitful to examine the various applications of these heterarchies, along with the distribution of knowledge among members and the dynamics of group interactions, as well as their mediated relationship to the overall system in which they function. The effects of having one individual manager accountable for specified results might be examined in terms of its relationship to group autonomy. It might be supposed that this delegation of individual responsibility could interfere with group autonomy and egalitarian participation; However, there was a remarkable lack of evidence to support this conjecture in the companies studied. The dynamics of these new relationships and roles should prove fascinating, and are worthy of future study, as are the variables which enhance or impede their effectiveness in practice, and the underlying belief systems which support or interfere with effectiveness of these structures.

The relationships between the ECCO system approach and other modeling and problem-solving methodologies should be explored for potential synergy, combination and enhancement. Most obvious will be the relationships with other systems methodologies. Of particular interest will be the potential use of ECCO system criteria with standard design and implementation strategies for Total Quality Management such as the Deming model, Juran's approach via Statistical Process Control, and various implementations being promoted by independent management consultants. The ECCO system criteria have the virtue of being organized, theory based and relatively mechanical, in contrast to Deming's 14 points, or the general management philosophy approaches put forth thus far. The combination of potential generality of application and the specificity of form in regulatory processes should provide a focus currently missing from many otherwise potent approaches.

Finally, it will be worth exploring aspects of the theoretical cybernetics applications. It is possible that the principles of heterarchy and of meta-process/result based control stated in the present context might set off some new formal applications of these principles to complex systems with radically different content. In addition, further theoretical relationships might be uncovered which could profitably extend, in breadth or depth, aspects of ECCO system principles. It is hoped that this work can serve to open new channels of inquiry, both applied and theoretical, in the service of making systems which work.


Back to Table of Contents On to Selected Bibliography Back to Home Page