[SysOp note: The electronic version of this article contains no graphics or figures. --Tom Glenn, the TQM BBS, 301-585-1164.] ----------------------------------------------------Business Index & ASAP------ AUTHOR(s): Juran, J.M. TITLE(s): Developmental quality planning. illustration table chart National Productivity Review p287(14) Summer 1992 v11 n3 Under the widely used Taylor management system, factories are managed through scientific methods rather than by use of the empirical "rule of thumb" so widely prevalent in the days of the late nineteenth century when Frederick W. Taylor devised his system. An important barrier to use of scientific management was the limited education of the lower level of supervision and of the work force. A large part of the factory population was composed of recent immigrants who lacked literacy in English. Some were completely illiterate. In Taylor's view, supervisors and workers with such low levels of education were not qualified to plan how work should be done. Taylor's solution was to separate planning from execution. To apply this solution, Taylor created planning departments, staffed them with engineers, and gave them the responsibility to: * Develop scientific methods for doing work. * Establish goals for productivity. * Establish systems of rewards for meeting the goals. * Train the personnel in how to use the methods and thereby meet the goals. Stunningly successful in raising productivity, Taylor's system was widely adopted in the United States, and was probably the chief reason for the U.S. rise to the position of world leader in productivity. Although the Taylor system originated in the factory production departments, the concept of separating planning from execution was universal in nature and, hence, had potential application to other areas: production support services, offices operations, service industries. The concept spread, and much of it was still in place at the end of the 1980s. In the nearly one hundred years since Taylor created his revolution, much has happened to make obsolete the premises on which he based his concepts: * The concept of separating planning from execution rested on the premise that the supervisors and work force lacked the education needed to perform the planning activities. Meanwhile education levels have risen remarkably in most countries. Lack of education is no longer reason enough to deny responsibility for planning to most supervisors and to many workers. * In Taylor's time the balance of power between managers and the work force was heavily weighted against the workers. (Industrial unions were illegal restraints of trade.) Since then industrial unionism (or the threat of it) has profoundly changed that balance. * There also have been changes in the climate of social thinking. Even in Taylor's time there were protests against the "dehumanizing" of work. Recent decades have seen strong trends toward giving the work force a greater voice in broad decision making, even extending to the concept of having substantial ownership in the companies. At present there is a lack of agreement on what should replace the Taylor system. The trends are to: * Revise the anatomy of the processes to minimize the use of short-cycle, repetitive jobs and the associated "hand offs" from worker to worker. * Enlarge the scope of jobs horizontally to include multiple tasks and skills. * Enlarge the scope of jobs vertically to include multiple functions. * Create self-supervising worker teams to perform the enlarged jobs. * Restore much of the responsibility for process planning, including planning for quality, to the line personnel. * Establish self-control. * Confer a higher sense of "ownership." * Enlarge job responsibilities to include responsibility for making improvements. * Provide in-depth training to enable the lower levels to carry out those added responsibilities. QUALITY PLANNING FOR MICROPROCESSES Compared to macroprocesses, microprocesses - operational systems involving few tasks, usually carried out in a single functional department - are much more numerous and much narrower in scope. There are also profound differences in organization form and in relationship to the hierarchy, as set out in Figure 1. **TABULAR DATA OMITTED** Microprocesses are so numerous that the upper levels of companies cannot become involved with them individually. However, some companies have undertaken initiatives to deal with the microprocesses collectively. They design a broad methodology intended to help departmental supervisors replan their own microprocesses. The broad methodologies are launched under such names as Departmental Activity Analysis. They differ in content, all the way fro vague exhortations to specific actions to be taken in accordance with a Departmental Analysis Manual. That manual sometimes provides forms and data sheets, along with instructions on how to fill them out. The subject matter addressed by the manuals varies considerably from company to company. Some of the topics relate to parameters other than quality. The more usual topics that do relate to quality are listed on page 289, together with brief descriptions. Define the Departmental Mission: The supervisors are asked to address such questions as: * Why does this department exist? * What tasks does the department perform? * What purpose should the department be fulfilling? The result of this analysis is a mission statement, which sets out what tasks the department should and can do. In some cases the concept of self-control is brought in as a criterion of what the department can do. This concept is readily depicted by a TRIPROL**TM** diagram (see Figure 2) **TABULAR DATA OMITTED** The TRIPROL diagram is a basic model used to explain the roles played by any department as it carries out its assigned quality mission. There are three of these roles: * Customer. The department acquires various kinds of inputs, which are used in carrying out its mission. The department is, therefore, a customer of those who provide the inputs. * Processor. The department processes those inputs to produce its products. The department is, therefore, a processor. * Supplier. The department supplies its products to its customers. The department is, therefore, a supplier. Figure 3 exemplifies the triple role concept applied to the recruitment of employees and preparation of financial reports. **TABULAR DATA OMITTED** Some of the companies' training materials go to considerable lengths to explain this basic and useful concept. Note that the triple role concept is applicable to any organization entity: the company, a division, a function, a department, a project team, a nonsupervisory worker. Any organization unit (of one or more persons) that carries out a prescribed process can be called a processor team. Identify the Customers: All manuals stress the importance of identifying the customers. However, they seldom explain the Big Q concept of customers, what steps to take to identify customers, and how to use the flow diagrams to help discover who is affected. Some manuals fail to point out the presence of internal customers. Others adopt the simplistic concept of focusing on the next recipient as the customer. Still others focus on the users of the products but fail to point out that all who are affected are customers. Establish Customer Needs: The manuals put much emphasis on determining the needs of customers. Most go beyond the relatively simple concepts of meeting specifications and requirements, and pose searching questions such as: * How is the product used by customers? * How do customers view the relative importance of the various needs? * How do the perceptions of customers differ from the perceptions of the supplying department? Some of the manuals include forms to be used as questionnaires for securing the needed answers from customers. Other manuals propose that the supplying departments consult with customers to understand customer needs in depth. All manuals stress the importance of closing any gap between customers and supplier regarding the perceptions of customer needs. In some companies, the manual stipulates that the analysis should be followed by a written agreement with customers, including internal customers. Describe the Process: All manuals include instructions for describing the process. Typically the instructions ask for a description that includes: * A list of the tasks or elementary operations performed within the department. * The time consumed in performing each of the tasks (hours, person-days, etc.). * The needs served by each of these tasks. * The output from these tasks. (This consists of products - goods or services in their various forms. In some manuals this output is described as "value added" or "contribution.") Identify the Suppliers and the Needs from Suppliers: All manuals require identification of suppliers, both external and internal. This leads to listing what is supplied by these suppliers and what the associated needs are. Some manuals point out that the problem of determining needs from suppliers parallels that of determining customer needs. The difference is that in this case the supervisor is the customer rather than the supplier. As before, the manuals urge reaching an agreement with the suppliers on the needs of the process. Establish Measures: All manuals mention the need for measures of quality. It is implied that these needs extend to each of the triple roles: supplier, processor, and customer. Occasionally a manual requires the supervisor to "evaluate process capability." However, the manuals do not spell out the methodology of arriving at measures. Here, as in the manuals generally, much is left to be filled in by training. Establish Feedback: The manuals point out the need to establish a feedback flow for the major paths within the TRIPROL diagram: * From the customers back to the supervisor's process. * From the supervisor's process back to the suppliers. The paths of this feedback can be added to the TRIPROL diagram, as shown in Figure 4. But the manuals seldom discuss how to establish these feedbacks. Review Performance and Take Corrective Action: The manuals usually stipulate that the supervisor should: * Collect feedback data on quality performances. * Compare the performances with the quality needs (which were previously converted into specifications, standards, procedures, etc.). * Identify nonconformances, defects, etc. * Take corrective action. Notice that these elements deal with quality control by use of the feedback loop, rather than with quality planning. Here again, the manuals are sketchy in the "how to" for the prescribed actions, so much will depend on supplemental training. Conduct Quality Improvements: The manuals are generally sketchy in Their descriptions of the process for quality improvement. The segments that are covered differ from manual to manual. Specifics in certain manuals include: * Analyze the cost of poor quality. * Identify problems and opportunities for improvement. * Select projects to be tackled. * Implement the improvements. LESSONS LEARNED FROM THE COMPANY APPROACHES Companies that have initiated supervisory analysis of microprocesses have done so to: * Improve the microprocesses trough replanning. * Use the experience and creativity of the supervisors (and of the nonsupervisory personnel). * Provide those same supervisors with greater participation and with a greater sense of ownership. * Broaden the supervisors' understanding of their roles. * Take some steps toward converting experienced amateurs into professional quality planners. To meet these objectives, some companies have prepared a structured methodology, including forms to be filled out and training in how to approach the replanning. These methodologies have their strengths and weaknesses. The strengths include: * A focus on supervisory ownership of the microprocesses. * Adoption of much of the quality planning road map. * Provision of a structured, understandable procedure, including forms to help in the data preparation. * Extensive listing of the actions to be taken. * Provision of training to supplement the manual. The weaknesses include: * Focus on microprocesses without provision for coordination with other elements of the macroprocess. * Individual responsibility for a replanning activity that inherently requires extensive teamwork. * Definition of what actions to take, but with inadequate explanations of the methodology to be used to take those actions. * Training often focused on the manual instead of on carrying out an actual replanning project. * Overly complex paperwork in some cases. * Inadvertent emphasis on departmental performance at the expense of overall company performance. In light of these lessons learned, upper managers should determine whether to take an initiative to provide supervisors with a structured approach for analyzing microprocesses. All too often companies have gone into replanning their microprocesses on the assumption that it would solve the quality problems of their macroprocesses. Yet, replanning of the microprocesses seldom solves the major problems of the macroprocess. Here are some better approaches to take: * Start with the macroprocesses. In general, the major problems of the macroprocesses can be traced to planning deficiencies at the macro level. Hence, the first efforts at replanning should be directed to the top of the process hierarchy. * Include supervisors in the planning teams. Planning of macroprocesses requires multifunctional teams. There is opportunity to include supervisors of microprocesses on such teams. The associated training and experience will help those supervisors later when they tackle the replanning of their microprocesses. * Establish an infrastructure for creating project teams. For many microprocesses the responsible supervisor will need the help of a team to make the analysis and provide the remedies. It should not be left to the supervisors to create such teams by the methods of "beg, borrow, and steal." The upper managers should provide a channel through which the supervisors' proposals for setting up the needed teams can receive approval and legitimacy. * Include emphasis on projects relating to product salability. The manuals have generally emphasized projects for reducing the cost of poor quality. Although these are important, making the products salable has even higher priority. * Explain the distinction between sporadic and chronic quality problems The former are handled by the control process; the latter require a structured approach to quality improvement. * Focus on getting projects completed. The company manuals have focused on training - training related to actions to be taken. Such training is usually forgotten unless it is built around a project to be completed. The focus should instead be on completing a project, and the training should be considered an aid to that end. * Provide training in "how to do it." The company manuals have generally limited themselves to defining what to do, but with little on how to do it. The manuals should contain (or be supplemented by) information on team building, the methodology and tools for analysis, the provision of remedies, the elements of self-control, etc. * Provide facilitators/trainers. Needed training is seldom acquired through self-study. There is need for facilitators/trainers who should first undergo intensive training in the subject matter and in how to train. * Include an example all worked out. Supervisors who have undergone training often suggest that the training materials provide a case example all worked out. * Test the extent of paperwork before issuing the manual. Some company initiatives have failed because of the amount and complexity of paperwork demanded from the supervisors. A massive movement is under way to revise the system of separating planning from execution. The new concept is to delegate more and more of the planning to the supervisors and the nonsupervisory personnel. This new concept is inherently sound, and the movement is destined to spread throughout the economy. We are still in the early stages of this massive movement. A number of substantial companies have tried to apply this concept to their operations, but their efforts during the 1980s did not result in establishing a widely accepted approach. This failure to secure wide acceptance seems to stem from the methods of application rather than from deficiencies in the basic concept. The methods of application can be improved, however. The feedback from the applications made to date have provided specific suggestions on what to do to make the improvements. This massive movement is irreversible. The companies will continue their efforts to delegate more and more of the planning to the lower levels. As they do, they will improve their methods of application and will secure results that will stimulate more and more companies to do the same. WHO SHOULD ANALYZE THE MICROPROCESS? Analyzing the microprocess involves some time-consuming tasks: filling out questionnaires, contacting those who are affected, preparing the flow diagrams and spreadsheets, collecting and summing up the available data, tracking down other needed data, and more. Various options are available for getting this work done. Full-Time Analysts: A full-time systems analyst, quality engineer, procedures analyst, industrial engineer, or other type of analyst is assigned to conduct the analysis. The analyst: * "Makes the rounds" to interview the cognizant people - managers, supervisors, workers. * Prepares the flow diagrams and other aids to collecting and presenting information. * Collects and summarizes the pertinent data. * Prepares a report, including recommendations for revisions. For the line supervisor, a full-time analyst helps solve the problem of finding the time to do the analysis. However, this approach has deficiencies, which some companies are no longer willing to accept: * The approach retains much of the concept of separating planning from execution. This concept is increasingly out of date at departmental levels. * The analysts exhibit the biases inherent in their culture. These biases then enter their recommendations. * Use of full-time analysts tend to assign "ownership" of the process to the analysts, rather than to the line personnel. * Lack of participation in the analysis reduces the line supervisors' willingness to support the resulting recommendations. The Departmental Supervisor: In this approach the supervisor of the microprocess is given a "do-it-yourself" responsibility to get the analysis done. The supervisors are assisted by a manual, such as a "Departmental Quality Analysis" manual. Teams of Supervisors: Under this concept a team is organized, consisting of the supervisor of a microprocess plus other supervisors who are customers and suppliers. This team is assigned to do the analysis and replanning. This assignment is preceded by training in how to plan for quality. Use of these supervisory teams has the potential to: * Arm the supervisors with the tools and skills of quality planning. * Increase the supervisors' sense of participation in quality planning. * Increase the supervisors' sense of "ownership" of the process. * Minimize the effect of biases of full-time planners. * Increase the willingness of customers and suppliers to support the recommendations of the analysis. The big disadvantage of using supervisory teams is the time it takes. Multiple people (teams) must do what could be done by one trained individual. In addition, this added work is often superimposed on the time of people who are already carrying a full-time load. Teams that Include Nonsupervisors: The massive movement to revise the system of separating planning from execution includes a movement to enable workers to participate in decisions that affect their jobs. A potential form of such participation is membership on teams set up to replan processes. The premises are that: * All workers have intimate knowledge of job conditions and are therefore able to make useful contributions to such teams. * These contributions can include identification of problems, theories of causes, and suggestions for remedy. * Many workers want to contribute. * Such contributions increase worker morale, provide a sense of ownership, and improve management-worker relations generally. To date, organized methods to provide for worker participation have consisted largely of: * Provision for individual contribution through methods such as suggestion systems. * Provision for group contribution by teams of workers through methods such as quality circles (QC). * Redesign of jobs. These methods have been widely tested, so that their strengths and weaknesses are well known. What has not been widely tested is the use of "mixed" teams composed of supervisors and workers. During the 1980s some companies did experiment with such teams. Those experiments were comparatively few in number, and the results did not lead to a clear consensus as to the effectiveness of this type of team structure. As of the early 1990s the jury was still out on mixed teams. When done systematically, the result of the analysis is an information package consisting of: * The conclusions reached. * Proposals (recommendations) relative to products, processes, inputs, etc. * Supporting information: flow diagram, spreadsheets, data, etc. The proposals are generally presented in the standard format used in the company. Some of the proposals concern matters over which the microprocess has full jurisdiction. In such cases the department supervisor can adopt them. Other proposals require concurrence by or action from other departments. Such proposals then require follow-up until (a) the recommended action is taken, (b) the recommendation is modified, or (c) there is a decision by authorities not to adopt the proposals. WORKERS AND THE TRIPLE ROLE At the worker level, the basic quality planning model is the TRIPROL diagram, Figure 2. Each worker plays all the roles of the TRIPROL model: customer, processor, and supplier. In many job situations, however, the workers have not thought of their jobs as consisting of those three roles. For example, an internal survey conducted by one public utility found that only 25 percent of the workers recognized the fact that they had customers. Such findings are to be expected in situations where planning has been cleanly separated from operations. It is common in such cases for workers to perceive their role as a matter of following procedures and meeting the goals. With that perception, it is understandable if workers conclude that: * The inputs (procedures, goals, facilities, data, materials) are determined and fixed during the planning. * The role of the worker is to conduct the process so as to conform to procedures and specified goals. * The product is "conformance." * The customer is the supervisor/company. If we redraw the TRIPROL diagram according to these perceptions, the result is something like Figure 5. This figure shows that the planners are perceived as the prime suppliers for the planning inputs, consisting of specifications, procedures, and quality goals. These same planning inputs are regarded as rigid and sacred - only the planners may change them. Other suppliers provide subsequent and continuing inputs: the data, materials, and so on, which become the grist for the processing mill. These inputs do vary, and the workers have a degree of responsibility to cope with these variations, as well as with variations inherent in the processing facility. Turning to the output side of the figure, the product is perceived as consisting of "conformance" to those specifications, procedures, and quality goals. The customer includes the supervisor and/or the company, but not subsequent users. These perceptions are reinforced if workers are kept out of the mainstream of feedback from customers to the process. The perceptions are further reinforced if the workers are kept out of the mainstream of feedback from the process to suppliers. To get its workers to participate in quality planning, an organization must create a methodology and provide the opportunity. The Statistical Process Control (SPC) movement has trained many workers to apply basic statistical tools to strengthen quality controls, while the QC concept has enabled many teams of workers to make improvements on job related matters. Another approach to worker participation in quality planning is to redesign jobs in ways that bring the worker closer to the status of artisan. The artisan is solidly involved in quality planning through (a) direct exposure to the needs of various customers and (b) being his own customer, over and over again. The opportunity for such redesign of jobs is largely traceable to the rise in education levels since the days of Frederick W. Taylor. In most cases it is not feasible to go to a full artisan concept, the damage to such parameters as cost and productivity is too great. So there is much groping and some experimentation. The job designs undergoing tests have included: * Job Enlargement-horizontal. Under this concept, jobs of a repetitive short-cycle content are converted into jobs of broader scope. A factory assembly line is redesigned to enable each worker to perform multiple tasks - even to assemble units of product completely. Similar job enlargements have been carried out in the office. In a telephone directory case, the work had been done as a procession of twenty-one different short-cycle tasks. After redesign, each worker performed all twenty-one tasks to produce a complete directory. In such job redesigns the workers become their own customers, over and over again, leading to easier identification of deficiencies in quality planning. * Job Enlargement - Vertical. In this approach the worker is assigned multiple functions. For example, a production worker may be made responsible for material supply, tool maintenance, judgment of product conformance, and so on. Such job designs bring the worker into quality planning of a multifunctional nature. In one company, the assigned responsibilities of production workers include many functions, with associated planning: * Self-Supervising Teams. This is a form of job enlargement that is both horizontal and vertical. An added significant feature is that the teams are largely self-supervising. The team decides which workers are to perform which operations. The team also takes on multiple functions. The need for quality planning is extensive, and the team participates actively in such planning. The term "new work systems" is used in some companies to describe a planning approach that focuses on creating self-regulating work systems involving little or no management and very few specialists. Under this concept, workers are paid according to skills attained rather than according to job classification or seniority. Workers within these systems schedule their own work times, shift assignments, job assignments, and vacations. Within these systems are ongoing teams, which review the social structure as well as the technology. For example, in one chemicals company a design team of sixteen workers from a 180-person department met weekly to design new or improved production systems. Their technical review covered ten key unit processes (microprocesses). They identified seventy-two possible variances in these areas. They then built an impact matrix to determine the effect of these variances downstream. They developed flow diagrams and decision trees for control of the key variances. All this resulted in twelve recommendations for changes in manufacturing technology. Their social review involved questionnaires and interviews for each of the 180 employees. These resulted in five recommendations for changes in job design or organization design. OBSTACLES TO WORKER PARTICIPATION IN QUALITY PLANNING The obstacles to bringing workers actively into the quality planning process include: * Managers who remain concerned about losing "prerogatives" and about the relative value of employee time spent on projects versus time spent producing the product. * Staff specialists who face competition in planning and analysis - a threat to the near - monopoly they have enjoyed under the Taylor system. * Workers who are concerned with threats to their job security and with extra rewards for project work. * Unions that are wary of shifting employee loyalty from the union to the company. Overcoming these obstacles is not simply done by logical reasoning. Results from field tests of the concept of worker participation in quality planning have, however, proved decisive in winning converts to the cause of involving workers in quality planning at the department level.