Analysis methods for production management

Thread Content

The various analysis methods used in management science for production and business management are nothing more than the application of various models based on scientific methods, with these models being used to represent the entire system under study or certain of its components. When analyzing problems in various fields, it is first necessary to determine the system boundaries of the study in order to define its scope. The guiding principle for defining the scope is to accurately determine which factors or variables may have an impact on the system under study. Generally speaking, the broader the boundaries or scope of a problem, the less likely suboptimal outcomes are to occur. 　　Next is the construction of the model. When constructing a model, it should be adapted to the actual production conditions; minor factors should be excluded, and the factors that have an impact on the production process should be analyzed in detail. At the same time, the relationship between controllable and uncontrollable factors must be taken into account in order to determine which model to use. The selection of a model is primarily determined by the relationships and interactions among factors. In analytical methods, it is necessary to determine the criteria for measuring efficiency and establish an effective set of standards to assess the effectiveness of various alternative options in production activities. The metrics for these aspects can include profit, contribution, total cost, incremental cost, machine downtime, machine utilization, labor cost, labor utilization, number of units produced, process time, and so on. 　　All models that use quantitative methods to study production problems can be summarized by one formula: E=f(*, yj). Here, E represents efficiency, f denotes the functional relationship, x represents the controllable variable, and y represents the uncontrollable variable. Controllable variables are those factors that can be manipulated and adjusted to a large extent according to the manager’s wishes. Uncontrollable variables refer to those factors that managers cannot control, or at least are not within the scope of the given problem. The meaning of this formula is that E (efficiency) can be expressed as a function of those variables that define the system. Once the model is established, E can be used as a measure of the efficiency of various alternative options in production activities. Based on this analysis, different alternatives can be generated, and these alternatives can then be evaluated. 　　The listed analysis methods mainly include: cost analysis, linear programming, queueing models, simulation techniques, statistical analysis, network planning models, heuristic models, computer-based search methods, graphical and image analysis, etc. These methods have corresponding applications in various aspects of production systems. 　　 Cost analysis Cost analysis is the most commonly used analysis method. This method is based on knowledge of the characteristics of various cost factors and comes in many forms. It is not a mere accumulation of accounting figures, but rather a numerical representation of the business’s performance. From the relevant figures, managers can obtain valuable information. Managers are not concerned with abstract costs; what interests them are the specific cost changes associated with the various alternative options they are considering. The basic method for cost analysis is the break-even analysis, which involves using the differences in how various costs change as the scale of operations changes to conduct the analysis. 　　The incremental cost analysis method is one of the most valuable simple analysis methods. It is used solely to study the costs affected by the policies or actions that might be adopted. Berfa pointed out that analyzing costs is not about calculating the total operating cost of each alternative option, but rather about examining the specific costs that differ when comparing different options. These costs mainly refer to inventory costs, adjusted labor costs, overtime fees, and outsourcing costs. Incremental cost analysis is widely used in various fields of production system analysis, and is commonly found in linear programming and queueing analysis models. 　　Linear programming: The essence of linear programming is optimization, that is, a mathematical method for finding the optimal solution based on a certain criterion, while satisfying specified constraints. Linear programming is a very important general-purpose model used primarily to address the issue of how to allocate limited resources in an efficient manner, thereby achieving maximum benefits within given constraints. Linear programming is widely applied in various fields such as industry, agriculture, management, and **science; it is one of the most effective tools used by modern managers and decision-makers. Linear programming can be used to address issues such as determining the optimal mix of multiple varieties in production, inventory control, raw material supply, and so on. In practical applications, linear programming often presents difficulties in finding solutions; to address this, the \"simplex method\" is generally used. 　　Queueing models Many production problems involve queueing to some extent. As long as there is random distribution in the production process, queues will definitely form. Various types of inventory are essentially buffers for queuing. Perfect equilibrium does not exist in reality. In this type of problem, people, components, or machines that require a certain service will be encountered at random intervals. The activities required to provide such a service often take varying lengths of time. Under certain arrival and service rates, mathematical methods can be used to calculate and arrange queueing problems. In contemporary times, queueing analysis is widely applied to service systems such as communication systems, traffic systems, production systems, and computer management systems. Queueing theory provides a mathematical approach to predicting the approximate length of a queue, the expected waiting time, as well as other relevant key metrics, including queue layout planning, priority service handling, queue cost control, and the relationship between queue length and the occurrence of delays. Having this information enables people to address related random distribution problems more effectively and make informed decisions. 　　Simulation technology　Simulation technology for production management problems is a rapidly growing technology. Although the basic concepts underlying simulation have existed for a long time, the rapid development of simulation technology has actually been driven by the advancement of high-speed computers. This method involves using mathematical models to conduct simulation experiments, processing relevant data with computers, and using the selected efficiency standard E as a measure to observe and examine the performance of various variables within the model. This model is experimental and may not necessarily produce the optimal answer. The advantage of simulation methods is that they allow for comparisons among various alternative solutions; in essence, it is a systematic approach to solving complex problems through repeated trials. 　　Statistical analysis Statistical analysis provides a set of methodological frameworks for the accurate processing of data. It can not only draw various conclusions based on the established predictive models, but also estimate the extent of the risk that predictions might be incorrect. Statistical analysis is often used in hypothesis testing, allowing us to address the significant variations in measured values of factors or variables within a system, factors or variables that may determine the scope of that system. By using statistical inference methods, conclusions can be drawn regarding the problems in related systems, and these conclusions can be quite accurate. In production and business management, statistical analysis methods have their own wide range of independent application areas. As a general tool for analytical methods, statistical principles can often assist in the application of other analytical techniques and contribute to overall analytical work. 　　Network planning model: After World War II, research and development activities as well as other large-scale one-time engineering projects became increasingly important in economic activities. In particular, the sheer scale and complexity of missile engineering and space programs necessitated special methods for engineering planning, scheduling, and control. This led to the emergence of network planning. The basic principle of network planning is to plan the tasks that need to be completed in the form of a network, including all the events involved in these tasks within that network. The arrangement of these events should be based on the sequence of construction activities and the interdependencies between different stages. Based on the network planning model, specific figures for the progress of tasks can be calculated, enabling managers to have a clear understanding of the planned schedule for these tasks and to allocate time more flexibly and effectively. The form adopted by network planning technology is unique, especially the concept of the \"critical path\", load balancing, the minimum cost method, and the allocation of limited resources. These interrelated concepts can provide a sound basis for the management of engineering projects. There are two specific methods in network planning models: one is the critical path method, and the other is the program evaluation method. 　　The Critical Path Method (CPM) was originally an internal planning method developed by DuPont in the 1950s; it was used for the planning and management of maintenance projects in chemical plants, and later became widely applied in many engineering projects. Around the same time, the U.S. Navy was also working to develop a method for planning and managing the development and production of the Polaris missile; this project required planning and oversight of nearly three thousand contracted entities, making the scale of the task quite considerable. As a result, the Program Evaluation and Review Technique (PERT) was developed. DuPont used CPM to reduce the time required for maintenance projects at its Louisville plant from 125 hours to 78 hours; the U.S. Navy utilized PERT to shorten the development time of the Polaris missile by two years compared to the original schedule. Although these two methods differ in detail, the principles on which they are based are similar. The difference between the two precisely illustrates how the practical needs of management influence technical methods. The critical path method was developed from maintenance engineering tasks that involve extensive practical experience, whereby the activity times for various tasks are known and the times along the path are definite. The program evaluation and review technique, on the other hand, originated in environments of exploration, research, and development; as a result, the activity times for various tasks there are highly uncertain, which is why probability methods are used to determine the expected times along the path. There are certain differences between the two regarding the details of how to create arrow diagrams. The network system created by the critical path method is slightly simpler; it uses nodes to represent event activities, and arrows to indicate the sequence of various events in the project. The Program Evaluation and Review Technique requires variables to schedule time. 　　The network planning model summarizes a lot of important information in a compact form. From the network plan, it is easy to calculate things such as the earliest and latest completion times, the float available in the activity schedule, the critical path, and so on. Given the statistical data on the activity network, critical path, and schedule calculations, a specific plan can be provided to managers for managing the project. Due to the time differences inherent in the event activities themselves, as well as those in the project completion dates, flexibility in planning the schedule is possible, allowing for different schedule options to be developed. Then, the usage can be compared from the perspective of resource load balance, and the value of a solution can also be assessed based on the level of activity costs. Different activities respond differently to changes in resource utilization; some may show no reaction to such changes, and in fact, slowing down the pace might result in higher costs than proceeding at a normal pace. Since different activities have different “cost-time” characteristics, it is possible to weigh the costs. If there is a shortage of key resources, due to their scarcity, it is necessary to review the plans from the perspective of making appropriate use of these limited resources. Time differences can be utilized to reduce resource consumption, and a trade-off between time and resource usage can even be made by extending the project timeline. 　　Heuristic model: The term “heuristic” itself implies the ability to guide managers in finding answers. In terms of managerial significance, heuristic models refer to guiding principles used for decision-making. Perhaps these guiding principles are not the best, but they are consistent when applied by people and effective, allowing the avoidance of more complex problem-solving procedures. Here, Berfa clearly draws on Simon’s theory of satisficing decision-making; he emphasizes that there are many issues for which we either lack the time or the interest to seek more thorough answers, but the existing principles are sufficient to help us find workable solutions. Perhaps it’s not absolutely correct, but this simple rule is the most applicable. Most of the problems faced by managers are complex; conducting a thorough and precise analysis proves difficult, and it is hard to solve them using mathematical methods. Yet answers must still be found, and in such cases, the logical framework established by business principles represents the best approach available. A so-called heuristic model is precisely such a set of logical and coherent rules. In a sense, heuristic methods are the oldest forms of thinking in management. Through this method, the amount of work required for exploration is reduced, at the cost of potentially giving up the best solution. In business management, this method is widely used in areas such as the balancing of assembly lines, equipment layout, scheduling of workshop operations, selection of warehouse locations, inventory control, and scheduling of one-time projects. 　　Computer-based exploration for solutions For some extremely complex problems, using computers to explore possible solutions can be an effective approach. The development of computer technology has promoted the application of heuristic models; computers can be used to sequentially examine a finite set of feasible trial solutions for certain criterion functions. By specifying the value of each independent variable, calculating the criterion function, and recording the relevant results, an experimental evaluation value can be obtained. Each test evaluation value is compared with the best value obtained previously; if it proves to be significantly superior, it is adopted in place of the previous best value, and this process continues until no further optimal solution can be found. This is the mountain-climbing style of step-by-step exploration. On this basis, the computer can, according to a predetermined operating procedure, print out the optimal combination of the identified independent variables. The advantage of the direct computer exploration method is that it establishes a criterion function model, which is not limited by linear mathematical forms and overcomes the constraint on the number of variables. In business management, computer-based search methods have been used to develop overall plans and task scheduling plans, as well as to address issues related to work allocation under limited resources. Computer-based exploration methods offer greater flexibility in business management; they do not require sophisticated model design or strict mathematical formulations, which allows for more freedom and enables cost models to be closer to reality. Therefore, Berfa believes that computer-based exploration methods for solving problems have an increasingly important role in real-world management. 　　Diagramming and Image Analysis Diagramming and image analysis are traditional methods used in production systems, with the most important form of such analysis being flowcharts that illustrate the sequence of activities or the timing of tasks.