SHOP FLOOR CONTROL
Shop floor control is concerned with Ihe release of production orders to the factory, monitoring and controlling the progress of the orders through the various work centers, and acquiring current information on the status of the orders. A typical shop flour control system consists of three phases: (1) order release, (2) order scheduling, and (3) order progress. The three phases and their connections to other functions in the production management system are pictured in Figure 26.9. In today's implementation of shop floor control, these phases are executed hy a com hi nation of computer and human resources, with a growing proportion accomplished by computer automated methods.
"he order release phase of shop floor control provides the documentation needed to process a production order through the factory. The collection of document is sometimes called the shop packet. It consists of: (1) the route sheet, which documents the process plan for the item to be produced: (2) material requisitions [0 draw the necessary raw materials from inventory: (3)job cards or other means to report direct labor time devoted to the order and to indicate progress uf the order through the factory; (4) move tickets to authorize the material handling personnel to transport parts between work centers in the factory if this kind of authorization is required; and (5) parts list, if required for assembly jobs. In rho operation of a conventional factory, which relies heavily on manual labor, these are paper documents that move with the production order and are used to track its progress through the shop. In a modern factory, automated identification and data capture technologies (Chapter 12) are used to monitor the status of production orders, thus rendering the paper documents (or at least some of them) unnecessary. We explore these factory data collection systems in Section 26.4.4.
The order release module is driven by two inputs, as indicated in Figure 26,9. The first is the authorization to produce that derives from the master schedule. This authorization proceeds through MRP which generates work orders with scheduling information. The second input to the order release module is the engineering and manufacturing data base which provides the product structure and process planning information needed to prepare the various documents that accompany the order through the shop.
The order scheduling module follows directly from the order release module and assigns the production orders to the various work centers in the plant. In effect, order scheduling executes the dispatching function in PPC. The order scheduling module prepares a dispatch list, which indicates which production orders should be accomplished at the various work centers. It also provides information about relative priorities of the different jobs, for example, by showing due dates for each job. In current shop floor control practice, the dispatch list guides the shop foreman in making Work assignments and allocating resources to different jobs so that the master schedule can best be achieved.
The order scheduling module in shop floor control is intended to solve two problems in production control: (1) machine loading and (2) job sequencing. To schedule a given
set of production orders or jobs in the factory, the orders must first be assigned to work centers. Allocating orders to work centers is referred to as machine loading. The term shop
loading is also used, which refers to the loading of all machines in the plant. Since the total number of production orders usually exceeds the number of work centers, each work center will have a queue of orders waiting to be processed. The remaining question is: In what sequence should these jobs be processed?
Answering this question is the problem in job sequencing. Job sequencing involves determining the sequence in which the jobs will be processed through a given work center.!o determine this sequence, priorities are established among the jobs in the queue, and the jobs are processed in the order of their relative priorities, Priority control is a term used in production control to denote the function that maintains the appropriate priority
levels for the various production orders in the shop. As indicated in Figure 26.9, priority control information is an important input in the order scheduling module. Some of the dispatching rules used to establish priorities for production orders in the plant include:'
First-come-first serve. Jobs are processed in the order in which they arrive at the machine. One might argue that this rule is the most fair.
Earliest due date. Orders with earlier due dales are given higher priorities.
Shortest processing time. Orders with shorter processing times are given higher priorities.
Least slack time. Slack lime is defined as the difference between the time remaining until due date and the process time remaining. Orders with the least slack in their schedule are given higher priorities.
Critical ratio. The critical ratio is defined as the ratio of the time remaining until due date divided by the process time remaining. Orders with the lowest critical ratio are given higher priorities.
When an order is completed atone work center, it enters the queue at the next machine in its process routing. That is, the order becomes part of the machine loading for the next work center, and priority control is utilized to determine the sequence of processing among the jobs at that machine.
The relative priorities of the different orders may change over time. Reasons behind these changes include: (1) lower or higher than expected demand for certain products,
(2) equipment breakdowns that cause delays in production, (3) cancellation of an order by a customer, and (4) defective raw materials that delay an order. The priority control function reviews the relative priorities of the orders and adjusts the dispatch list accordingly.
The order progress module in shop floor control monitors the status of the various orders in the plant, WIP, and other characteristics that indicate the progress and performance of production. The function of the order progress module is to provide information that is useful in managing the factory based on data collected from the factory. The information presented to production management is often summarized in the form of reports, such as the following:
Work order status reports. These reports indicate the status of production orders. Typical information in the report includes the current work center where each order is located, processing hours remaining before completion of each order, whether the job is on-time or behind schedule, and priority level.
Progress reports, A progress report is used to report performance of the shop during a certain time period (e.g., week or month in the master schedule).!! provides information on how many orders were completed during the period, how many orders should have been completed during the period but were not, and so forth.
Exception reports; An exception report indicates the deviations from the production schedule (c.g .. overdue jobs]. and similar exception information
Factory Data Collection System
There arc a variety of techniques used to collect data from the factory floor. These techniques range from clerical methods which require workers to fill out paper forms that are later compiled, to fully automated methods, that require no human participation. The factory data collection system (FDC system} consists of the various paper documents, terminals, and automated devices located throughout the plant for collecting data on shop floor operations. plus the means for compiling and processing the data. The factory data collection system serves as an input to the order progress module in shop Floor control. as illustrated in Figure 26.9. It iv also an input to priority control, which affects order scheduling. Examples of the types of data on factory operations collected by the FOC system include: piece counts completed at a certain work center. direct labor time expended on each order. parts that are scrapped. parts requiring rework, and equipment down-time. The data collection system can also include the time clocks used hy employees to punch in and out of work.
The ultimate purpose of the factory data collection system is twofold (1) (0 supply status and performance data to the shop floor control system and (2) to provide current information to production foremen, plant management. and production control personnel To accomplish this purpose. the factory data collection system must input data to the plant computer system. In current Cl M technology. this is done using an online mode, in which the data arc entered directly into the plant computer system and are immediately available to the order progress module, The advantage of online data collection is that the data file representing the status of the shop can be kept current at all times. As changes in order progress are reported. these changes arc immediately incorporated into the shop status file. Personnel with a need to know can access this status in real-time and be confident that they have the most up-to-date information on which to base any decisions. Even though a modern FOC system is largely computerized. paper documents are still used in factory operations, and our coverage includes both manual (clerical) and automated systems.
Manua,1 (Clerical) Data Input Techniques. Manually oriented techniques of factory data collection are those in which the production workers must read from and fill out paper forms indicating order progress data. The forms are subsequently turned in and compiled, using a combination of clerical and computerized methods. The paper forms include:
Job traveler; This is a log sheet that travels with the shop packet through the factory. Workers", ho spend time on the order are required 10 record their times on the log sheet along with other data such 'l'i the date, piece counts, defects. and so forth. The job traveler becomes the chronological record of the processing of the order. The problem with this method is irs inherent incompatibility with the principles of real-time data collection. Since the job traveler moves with the order, it is not readily available for compiling current order progress.
Employee time sheets. In the typical operation of this method. a daily time sheet is prepared for each worker, and the worker must fill out the form to indicate work that he/she accomplished during the day. Typical data entered on the form include order number, operation number on the route sheet, number of pieces completed during the day, and time spent. Some of these data are taken from information contained 1ll the documents traveling with the order (e.g., typical documents traveling with the order include one or more engineering drawings and route sheets). The time sheet is turned in daily, and order progress information is compiled (usually by clerical staff).
Operation tear strips. With this technique, the traveling documents include a set of preprinted tear strips that can be easily separated from the shop packet. The preprinted data on each tear strip includes order number and route sheet details. When a worker finishes an operation or at the end of the shift, one of the tear strips is torn off. piece count and trme data are recorded by the worker. and the form is turned in to report order progress.
Prepunched cards. This is essentially the same technique as the tear strip method. except that prepunched computer cards are included with the shop packet instead of tear strips. The prepunched cards contain the same type of order data, and the workers must write the same kind of production data onto the card, The difference in the use of pre punched cards is that in compiling the daily order progress, mechanized data processing procedures can be used to record some of the data.
There arc problems with all of these manually oriented data collection procedures. Thev all rely on the cooperation and clerical accuracy of factory workers to record data onto a paper document. There are invariably errors in this kind of procedure. Error rates associated with handwritten entry of data average about 3% (one error out of 30 data entries). Some of the errors can be detected by the clerical staff that compiles the order progress records. Examples of detectable errors include: wrong dates, incorrect order numbers (the clerical staff knows which orders are in the factory, and they can usually figure out when an erroneous order number has been entered by a worker), and incorrect operation numbers on the route sheet. (If the worker enters a certain operation number, but the preceding operation number has not been started, then an error has been made.) Other errors are more difficult to identify. If a worker enters a piece count of 150 pieces that represents the work completed in one shift when the batch size is 250 parts, this is difficult for the clerical staff to verify. If a different worker on the following day completes the batch and also enters a piece count of 150, then it is obvious that one of Ihe workers overstated his/her production, but which one? Maybe both.
Another problem is the delay in submitting tile order progress data for compilation. There is a time lapse in each of the methods between when events occur in the shop and when the paper data representing those events are submitted. The job traveler method is the worst offender in this regard. Here the data might not be compiled until the order has been completed, too late to take any corrective action. This method is of little value in a shop floor control system. The remaining manual methods suffer a one-day delay since the shop data are generally submitted at the end of the shift, and a summary compilation is nor available until the following day. In addition to the delay in submitting the order dahl, there is also a delay associated with compiling the data into useful reports. Depending on how the order progress procedures are organized, the compilation may add several days to the reporting cycle.
Automated and Semi Automated Data Collection Systems. Because of the problems associated with the manual/clerical procedures. techniques have been developed that use data collection terminals located in the factory. Data collection terminals require workers to input data relative to order progress using simple keypads or conventional alphanumeric keyboards. Data entered by keyboard are subject to error rates of around O.3%"(one error in 300 data entries), <III order uf magnitude improvement in data accuracy over handwritten entry. Also, error-checking routines can be incorporated into the entry procedures to detect syntax and certain other types of errors. Because of their widespread use in our society. PCs are becoming more and more common in the factory, both for collection of data and for presenting engineering and production data to shop personnel.
The data entry methods also include more automated input technologies such as optical bar code readers or magnetic card readers. Certain types of data such as order number. product identification, and operation sequence number can be entered with automated techniques using bar-coded or magnetized cards included with the shop documents (refer back to the barcoded route sheet in Figure 12.7).
Using either PCs or terminals that combine keypad entry with bar code technology, there are various configurations of data collection terminals that can be installed in the factory. These configurations include:
One centralized terminal In this arrangement there is a single terminal located centrally in the plant. This requires all workers to walk from their workstations to the central location when they must enter the data. If the plant is large, this becomes inconvenient. Also, use of the terminal tends to increase at time of shift change, resulting in significant lost time for the workers.
Satellite terminals. In this configuration, there are multiple data collection terminals located throughout the plant. The number and locations are designed to strike a balance between minimizing the investment cost in terminals and maximizing the convenience of the plant workers.
Workstation terminals. The most convenient arrangement for workers is to have a data collection terminal available at each workstation. This minimizes the time lost in walking to satellite terminals or a single central terminal. Although the investment cost of this configuration is the greatest, it may he justified when the number of data transactions is relatively large and when the terminals are also designed to collect certain data automatically.
The trend in industry is toward more use of automation in factory data collection systems. Although the term "automation" is used,many of the techniques require the participation of human workers; hence, we have included "semi-automated" in the subtitle for this category of data collection system