Get best of the week

The source of production efficiency

How to achieve real efficiency of the production and logistics system


Peterkin S.V., CPIM



annotation


The struggle for "production efficiency" in the Russian Federation, for the vast majority of enterprises, is moving into a sluggish stage. This is explained by the fact that well-advertised methods, such as long and successfully used "there" (Lean (Lean Production), automation with ERP / APS / MES / ... systems), and "new-fangled", such as Industry 4.0 or do not give any performance improvements, either give a disproportionately small in comparison with the expended efforts or expectations, or simply do not start.

The main reason for this is a misunderstanding or, worse, a misunderstanding, to solve which problems, in which cases, or how, these concepts should / can be applied.

Accordingly, the path to improving efficiency must begin with an awareness of the model of operation of production and logistics systems ( PLC), the root causes affecting their theoretical effectiveness. Further, within the framework of the theoretical model, look for problem areas in a real plant and eliminate them with the tools intended for this.

Important comments

  1. “Efficiency” means the classical economic efficiency of the manufacturing business: more output with fewer operating costs and frozen (in stocks, equipment, labor) capital.
  2. The following theses are equally well applicable for any (discrete) production, any size and method of satisfying demand (“production-to-warehouse”, “production / assembly / configuration-to-order”, “development-to-order”), any method of organizing production (subject-closed sections and / or lines / posts / conveyor, physical or logical ...)
  3. By "any enterprises" we mean enterprises for which "efficiency", as defined above, is NOT just a number, which can be almost any, the main thing is that the annual report / presentation shows the heroic efforts to achieve it (effectiveness). These are productions for which “efficiency” is “service level”, “profit” for an order or for a period, “turnover”, “costs” for an order and for a period, etc. And failure to achieve which means sooner rather than later, "Departure" from the business ...

Problem and Solution


The manufacture of engineering / instrumentation products for an enterprise operating in a changing market is characterized by the following:

  • a) “run” for the market / customers (their requirements) or strive to get ahead of its (market) requirements. As a result, there is a constant parallel process of development- (improvement) -production of products, leading to a constant change in the Composition of Products (SI), the constant need for updating plans for the entire production and logistics chain (PLC, it is also the supply chain),

  • b) produce complex (often different each time) products


Under the "complex" refers to the manufacturing processes of products:

  1. with a large number of units (parts, materials, purchased products) in SI (“large” - begins with the first hundreds - sometimes tens - of positions in the design specification (SI),
  2. often - with a long (shift, days) production time of parts / assemblies,
  3. , : , ( /), . «» — « », – , 1 : , , : , (
), ( ) .


The indicated, as well as the desire / need to maximize autonomy of our own production from external conditions (suppliers / cooperators), especially in key nodes / repartitions, leads to the need to organize “complex”, multi-productive production (we do everything ourselves / almost everything / a lot). And for large enterprises - to the construction of a geographically distributed PLC, consisting, as an example, of the following nodes:

  • factory (s) for the production of blanks, parts of the "lower stages", plant (s) for the production of aggregates,
  • DSE production plant, (assembly) aggregate / medium, large units,
  • final assembly and inspection / testing.

The overall complexity of the production system leads to the complexity of the control system. The management system in this case is a system based on information management, which is transmitted in the form of a plan / factual information to all participants in the implementation of the order, on paper or using information systems.

From this context it follows that "complex production" is almost any plant, of several dozen people or more, producing products "more complicated than a (children's) bicycle."

This may seem surprising to some, but the root causes of the “difficulties” in managing “complex” PLCs that produce “complex” products have long been well known and are described in the literature [1, 2]. Also - the answer is known [3]. We give these reasons briefly.

The main reason for inefficiency


The main reason for inefficiency (expressed in a high level of stocks - increased production cycles, whip effect) of complex PLCs is a high response time to internal (inside PLC) or external (changes in demand / supply outside the PLC) changes. A high response time leads to fluctuations in (all) stocks throughout the PLC. Wherein:

  • the lower point of the inventory fluctuation curve leads to deficiencies in product components. And, as a result, to increase production time and / or increase operating costs (“burning supply” or production),
  • the high point of the fluctuation curve is toward increased stocks. What, in the case of WIP, is to increase production cycles (Little Law).




If this picture seems unrealistic to you, try looking at the profile of the change in the level of inter-operational stocks of DSE processed in partion.




“Speaking Russian ...” a high reaction time means that a new order / change of an old “arriving” at a plant does not immediately “throw in” into development / production / supply, changing plans of everyone and everything (except for the “frozen” period, of course) , and after some rather long period. As a rule, with a monthly general re-planning.

Similarly, when the internal conditions change (the supplier missed the deadline, there was a defect / revision in production), etc. the effect on the issue / orders is determined through what something, quite a long time. Once a month, when rescheduling. Or, for example, when an irritated customer calls. And if the customer is a VIP, the general director, in fact, works as a dispatcher ...

And all this quickly or gradually leads to what "warehouses / workshops are crammed to the ceiling, but there is always something missing in the assembly! "

Key factors affecting response time


1. Frequency of general re-planning

At the same time: an increase in the frequency of re-planning by 2 times leads to a decrease in inventory levels throughout the PLC by 1.7 times [3]. But, frequent rescheduling:

  • a) leads to an increase in the number of periods with a potential minimum of reserves - deficits,
  • b) requires, in the traditional approach, more administrative efforts to change / bring plans to the performers,
  • c) requires a different model for constructing SI, the actual planning algorithms, and a model for collecting and recording evidence.

2. The size of procured and, most importantly, produced lots.

At the same time: reduction in the size of launched lots:

  • a) leads to an increase in the number of periods with a potential minimum of reserves - deficits,
  • b) requires, in the traditional approach, more administrative effort (more “launches”),
  • c) up-to-date and quick stock information for the entire PLC,
  • d) requires optimization of readjustment times.


CONCLUSION: We often re-plan (ALL!) Production and deliveries, run smaller (better — strictly on order) batches — we get higher reaction rates, lower inventories (of materials, PKI, NZP), and we shorten the production cycle. But! More often, the potential transition of the stock level “through“ 0 ”, i.e. accurate process control is required.


Decision


An increase in the response speed is possible through a reduction in the reaction time of two main production flows:

a) “physical” (stocks, equipment, employees),
b) “information” (electronic or paper (“handwritten) information about the plans, status and fact of“ physical ”objects flow).





Good “tools” for reducing material flow changes at present are Lean methods. BUT! Provided that they are purposefully applied , and mainly for solving the task.

“Standardized operations”, “5C” (in fact, improving the ergonomics of workplaces), “Visualization”, “Supermarket”, “The flow of single products”, “Quick changeover” - if used correctly, they can significantly (several times!) Reduce the reaction time , and therefore - to increase the effectiveness of the PLC under consideration. Unfortunately, not many people know about this “side” of Lean. The vast majority of “lean manufacturing” implementations now are projects that are being planted from above, holdings, or projects initiated under the influence of “information noise”.

Performed without a deep understanding of both the essence of the methods / tools and the goal (reduction of reaction time). Such transformations, as a rule, begin with a big fanfare and end with dusty stands in workshops, bright signal lights and (“sucked out of the finger”) “huge” or really taking place, but local economic effects. Which can be shown to tourists or inspectors of the parent company.

Speeding up information flow

Obviously, for this it is logical to apply information technology. But, "as it turned out," the "simple" implementation of "control systems" implemented in traditional ERPs, plus, worse, MES / APS / SCM, etc., also does not give the expected effect. Because the approach is similar: “easy to implement.” Without understanding, “why” ... Ie without the goal of “reducing response time”, without understanding the industrial model / methodology (western!) of production embedded in these systems. And, as in the case of “working” with “material” flows through Lean, without understanding why and how these IT technologies (IT systems) should work in production / supply. This leads to long and difficult implementations, beginning and ending, as a rule, with automation of the simplest and not affecting the effectiveness of the MTS inventory management functions and accounting. Or - “automation of planning and production”,which is just a workflow automation of existing and imperfect planning processes.


Let's try to consider the area of ​​control of the "information" flow from the right point of view.

Responsive System Concept


Requirements


So, the right decision in the field of reducing the response time of the PLC information flow is a combination of two factors:

  1. Methodology of planning, management, monitoring, providing the shortest possible reaction time for a given production.
  2. Support for this methodology by an appropriate IT system.

The main purpose and role of the IT system is:

  1. Support for an efficient (fast) production model.
  2. Fast, frequent and correct (within the chosen production model) planning and re-planning.
  3. Providing controls (monitoring) and rapid impact ( Control Tower concept ).

The system (the System is a methodology + IT system), built on such principles, is hereinafter referred to as the JMP - Planning and Monitoring System (Production and Supply).

SPM concept


Conceptually, the “four pillars” of SPM are as follows.

  1. The classic SCM approach (as well as methods and computational algorithms), namely, modeling as a supply chain of a plant (each workshop is a supplier and / or consumer) and its environment (suppliers and cooperators), fast planning at different levels, from inter-factory cooperation to in-shop management.
  2. System dynamics, namely, its main postulate that the effectiveness of the production system depends primarily on the "response speed", i.e. the reaction time of the system to external and / or internal changes,
  3. Lean ( , ): /; lean ,
  4. On-line - , , , .

SPM, to satisfy the above statements, must satisfy the following requirements for construction and algorithms.

1. Support for the interconnected classic MRP-II system of plans . With details "the lower the higher." Which implies (and is practically confirmed), among other things, the uselessness of using MES without setting two "upper" levels of planning: demand management (balancing capacities, output, orders) with synchronization (orders-development-production-supply).



2. Strict custom planning. (BUT! Optional strict execution), where each “order” (and this is either a specific product or a batch of the same type of products for a specific release date), each product (s) of orders, incl. by "placer", for repairs - it is planned and managed in production separately, in terms of its composition, determined by the order number or a specific serial number of the product.




3. The availability of always 2 versions of plans: directive and settlement. Each product / each order is planned from the release date (from the date of transfer to the customer), “back” (by the time of the workshop / precinct production of assemblies or parts, up to a day) and “down”, throughout the entire structure of the product. And back. In a few iterations. Those. using SCM approaches and algorithms, but inside and outside the plant, for the entire production and logistics chain of the product. With the formation of both a “directive” (“as it should”) and a “calculated (“ as it turns out “) version of the plan.





“Fast” (on-line) communication of plans to the workshops and supplies (as well as to suppliers, cooperators, centers of competence, providing access to the PSD via the web interface) with daily “acceptance” from them of information about the actual progress of production / purchases. And then - a comparison of the fact with the plan, positionally, by quantity, by% of completion, with the receipt of a "calculated" version of the plan.



4. The KPI system , which stipulates the work of workshop engineers and design engineers “on order,” and “on time”.


Literature


  1. “Industrial Dynamics”, Jay Forrester, Productivity Press, 1961. As with the Theory of Constraints, the system dynamics model soon found its application not only in the field of inventory and production management, but also in management and economics. In the Russian translation - "Fundamentals of cybernetics of the enterprise (industrial dynamics)", publishing house "Progress", 1971.
  2. Factory Physics: Foundations of Manufacturing Management, third edition, 2008. 720pp. ISBN 978-0-07-282403-2.
  3. Maynard's Industrial Engineering Handbook, Fifth Edition. Chapter 76 "AN INTRODUCTION TO SUPPLY CHAIN ​​MANAGEMENT". Publisher: McGRAW-HILL: New York, Chicago, San Francisco, Lisbon, London Madrid, Mexico City, Milan, New Delhi, San Juan Seoul, Singapore, Sydney, Toronto. ISBN: 9780070411029

More articles:


All Articles