Gaining insight into capacities, bottlenecks, performance measurement and Performance Efficiency at Unilever’s Sauce factory in Oss

(1)

Gaining insight into capacities, bottlenecks, performance measurement

and Performance Efficiency at Unilever’s Sauce factory in Oss

Annelies Roll

August 2006, Groningen

(2)

Thesis

The Need for Speed

Gaining insight into capacities, bottlenecks, performance measurement and Performance Efficiency at Unilever’s Sauce factory in Oss

Author: Annelies Roll

University: University of Groningen

Faculty of Management and Organization

Business Administration

Production- and Servicemanagement

Technology Management Discrete Technology

University Supervisors: Dr. G.C. Ruël Dr. J. Riezebos

Company: Unilever Bestfoods

Sourcing Unit Oss (Sauce factory) Molenweg 80, OSS

Company Supervisor: Ir. L.H. Hulshof

Publication: Groningen, 31-08-2006

© 2006 A. Roll. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior permission of the author.

(3)

EXECUTIVE SUMMARY

Organizations tend to focus more and more on efficiency of the production system and increasing the net output. Several concepts like Lean Manufacturing and Total Productive Maintenance1 are available to support the organizations. In the pursuit to increase efficiency at the Sauce factory of Unilever in Oss, management came to the conclusion that overall there was insufficient insight in the production. Several questions were important like; “What are our bottlenecks?”, “Are our norm rates still accurate?”, “How do we have to use our buffers?” and “What decreases our performance?”

Founded on these considerations, a research is conducted at the Sauce factory of Unilever in Oss. The objective of this research is to create more insight in the capabilities, the bottlenecks and the performance of the production system of Unilever’s Sauce factory in Oss, in order to improve the net production rate.

In order to create more insight this research is conducted from February until August 2006, focussing on the different production areas in the Sauce factory. Firstly the capabilities of the production system are thoroughly analyzed and mapped in order to identify the bottlenecks. Data is collected on all separated stages for all product types. The capacities of the stages are very product-dependent and are bounded by product constraints like quality and heating as well. The bottlenecks though, mainly concentrate around the heat exchanger and the palletizer. Based on these bottlenecks, the capacities corresponding with these bottlenecks are called the bottleneck rate. These rates restrict the maximum output of the production system. Subsequently, these bottleneck rates are compared with the current norm rates. For 85 percent of the products the current norm rates differ from these bottleneck rates. This leads to scheduling problems, non-transparent transfer prices and inaccurate standards for performance measurement. The way performance is measured also has a negative effect on the data that is collected; it is too human-dependent, measured on the wrong place in the line and not complete. The unsuitable data on performance have a negative influence on the performance. When the right performance improvement are not set up, a lot of unexplained losses occur and use of the buffers is unsuitable, losses of

1

(4)

Performance Efficiency will occur, while increasing the Performance Efficiency is one of the main targets of the Sauce factory. This research only focuses on these losses, changeovers, start up losses and quality losses are not discussed. Founded on the analysis above, there is a clear need for improvement. Improvement of the net production rate is possible by improving the capacity and reducing (the effect of) variation of the bottleneck. Important hereby is also the improvement of the performance measurement system in order to increase the reliability of the data on performance. For these improvements the following main question will be applied:

How do the bottlenecks, performance measurement and Performance Efficiency have to be improved at the Sauce factory Unilever SU Oss in order to increase the net production rate?

For the bottlenecks, several improvement options are proposed, of which the most important are: speeding up the heat exchanger and an eight tray gripper for the palletizer. The former will increase the overall capacity by about nine percent, which could save approximately 284 hours for the year 2006 (based on the contracted volume for 2006). The bottleneck will shift for 87 percent to the tray packer. The performance measurement system will be improved by implementing an automatic measurement system at the tray packer. This system will collect more accurate, less human-dependent data that can be used for analyzing the performance. Thereby the packaging area has to become leading in scheduling, improvement projects and importance of the area. The variation in production rate is caused by big breakdowns, minor stoppages and reduced speed. In order to reduce these losses, a structural long term approach is necessary. For the breakdowns the focus has to lay on failure analysis and human failures. For the other losses first accurate data has to be collected and root-causes have to be investigated. Besides this, the FIFO-buffer has to be used to intercept variation and a new buffer has to be placed upstream of the bottleneck in order to reduce the effect of variation on the bottleneck.

Through these improvements suggested above, the overall capacity and the Performance Efficiency will be increased so that the net output rate will increase. Merely based on the bottleneck improvements the Sauce factory will be able to produce 5 million jars a year extra, an increase of 8 percent in volume. Thanks to this increase of capacity the Sauce factory will be able to meet the targeted OQ increase of 4 percent.

(5)

PROLOGUE

For the past seven years I have been captivated by the rather abstract and theoretical world of Business Administration and Technology Management. Even though I have truly enjoyed my time at the lecture halls of the University of Groningen, I was delighted to finally be able to bring my accumulated knowledge into practice during my internship at the Unilever’s Sauce factory in Oss. Although the task at hand appeared daunting at first, I was pleased to learn that I could indeed make a difference in a company. By analyzing the capabilities of the production system I have gained a lot of new insights for the Sauce factory. Based on my analysis, I was able to provide them with recommendations that not only changed the body of thought from push into pull thinking, but also informed them about the impact of bottlenecks in the factory. For instance, one of my recommendations regarding bottlenecks will lead to a yearly increase of Operational Efficiency of over four percent when acted upon. But before this thesis actually commences I would like to express my gratitude to a few people that really made a difference during my internship and the process of writing this thesis. First of all I would like to thank all the people of the Sauce factory for taking me on a 6 month trip in the world of tomato paste, ‘Fryma’s’ and ‘Kartoffel mal Anders’. In particular, I would like to thank Frank de Kok and Jose Reijmersdal for all the fruitful discussions and support they continued to express. In addition, special gratitude goes out to Loes Hulshof, whose continuing support and never-ending enthusiasm for my thoughts and ideas made the internship so much more enjoyable, whilst honing my personal skills in the process. I think I could not have wished for a better company supervisor. Furthermore, I would like to thank my supervisors at the University of Groningen; Jan Riezebos for giving me all the support I needed with respect to content, and Gwenny Ruel who really understood the kind of aid and reassurance I required in writing this thesis. Gwenny, I am sincerely grateful for the many hours you invested in me; you truly have been a great mentor and supervisor. Thanks go out to my friends as well for reading my thesis and giving me recommendations for further improvements; especially Jop, my mainstay. Lastly, I would like to thank my parents for giving me the emotionally support as well as the financial means which proved so indispensable during the course of my studies.

Annelies Roll

(6)

INTRODUCTION

“The Need for Speed.” In a world with increasingly demanding consumers and highly competitive markets, being fast is crucial. Companies in the food processing industries in particular, strive to achieve a fast and efficient production system in order to stay on top of their game in a highly mature market. Still, “the more hurry, the less speed” was never more apt; as blindly speeding up the production rate might very well lead to losses in efficiency. Therefore, it is crucial for any company to be aware of the critical processes in its production chain and the overall capabilities of its system. In other words, where are one’s bottlenecks?

As a global producer of a broad range of sauces, the Unilever Sauce factory in Oss has ceaselessly been searching for ways to raise the efficiency of its production system and increase its net output. In recent years, Unilever has attempted to achieve a similar objective through the adoption of the concept of Total Productive Maintenance (TPM). Consequently, the Sauce factory has been striving to optimize the Performance Efficiency (PE) of its production environment by means of this Total Productive Maintenance concept as well. Still, despite their best efforts, the Sauce factory has stumbled across a number of problems which have prevented the full realization of the Performance Efficiency targets. Particularly, the only partial perception of its productive capabilities and the failure to correctly monitor its overall performance, have severely hampered the improvement of efficiency and net production for the Sauce factory. Apart from this, the production system has been subject to a high degree of variation in its production rate; a problem, which has to be addressed, in order to fully realize the Performance Efficiency objectives. Accordingly, this research will analyze the current production system and formulate recommendations in order to improve the Performance Efficiency and optimize the net production rate. The objective of this research is thus:

To gain more insight into the capabilities, the bottlenecks and the performance of the production system of Unilever’s Sauce factory in Oss, in order to improve the net production rate.

(9)

This report starts with an introduction to the company, the Sauce factory of Unilever in Oss. In this section the main products, process flow and characteristics are described. The second section will throw further light on the primary motives for this research through a summary of recent problems. In the next section these problems are analyzed and the research design is set up. Afterwards, in the fourth section, a theoretical framework for this research is constructed. In the fifth and sixth sections a conceptual framework is set up as well as the research methodology, based on this theoretical framework. Based on these frameworks the next four sections analyse the current production system. This analysis starts with a description of the main process flow in section seven. Furthermore the capacity and bottlenecks (section eight), performance measurement (section nine) and Performance Efficiency (section ten) are analyzed. Section eleven consists of the final conclusions of this analysis and a setup of the design phase, the final phase of this research. The next three sections consist of a redesign of the production system, with the aim of raising the overall production rate. Subsequently the bottleneck improvements (section twelve), a redesign of the performance measurement system (section thirteen) and recommendations in order to improve the Performance Efficiency (section fourteen), are passed in review. This research is concluded with the final conclusions and recommendations.

This is the public version of a confidential report; therefore, certain confidential details are removed.

(10)

1. OUTLINE OF THE COMPANY

This research is conducted at the Sauce factory in Oss, part of Unilever. This section consists of a description of the company Unilever in general, the Sourcing Unit Oss (SU Oss) and the Sauce factory, where the research takes place. Furthermore, this section provides a description the primary process and its production characteristics.

§1.1. Unilever Corporate

Unilever was officially established in 1930, its origin dates back to the year 1870. In those days Simon van den Bergh and Anton Jurgens both owned a butter trading company in Oss and decided to co-operate in order to start exporting butter to England. In 1927 they decided to unite themselves in the Margarine Union. In the following years other European companies, like Calvé Delft and Hartog, also decided to join them. In the meantime, in 1885, William Hesketh Lever, a groceries wholesaler, was starting a large soap company in England, which was called Lever Brothers. In 1917 he did not only produce soap, but also fish, ice cream, and preserved food. In 1930 the Margarine Union and Lever Brothers owned businesses in more than 40 countries. Due to their coinciding interests concerning the raw material market and their consuming market, they decided to merge into Unilever.

As a major player in the Fast Moving Consumer Goods – branch (FMCG), the Unilever brand is a very well known company throughout the world. The company has developed to one of the largest producers of food, with net sales of over 40 billion Euro and about 250.000 employees spread over 151 countries. At this moment two legal parent companies are maintained, Unilever NV (Netherlands) and Unilever PLC (United Kingdom), working through an equalization agreement and other contracts between the two companies. It was a full business merger, operating as a single business entity; both have their own companies. Unilever operates as a unified board; the boards of NV and PLC comprise the same directors. The boards have ultimate responsibility for the business as a whole. Unilever operates world-wide in two major product groups:

• home and personal care (e.g. Axe and Omo) • food (e.g. Magnum, Knorr and Bertolli)

(11)

Unilever has developed a mission for the company valid for all Unilever organizations that operate in supply chain and sales/marketing world-wide;

‘Unilever’s mission is to add Vitality to life. We meet everyday needs for nutrition, hygiene and personal care with brands that help people feel good, look good, and get more out of life.’ (www.unilever.com)

The global group is organized as one company world wide with a divisional structure per continent (see appendix A), and each continent has a supply chain division while sales and marketing are country responsibilities. The SU Oss is part of the European supply chain, consisting of four main parts, interacting through multiple leads, reporting lines and a team structure (see fig. 1.1).

Fig. 1.1 Structure supply chain Division Europe

The European supply chain is divided into eight categories. SU Oss is part of the savory category which consists of 22 factories. These factories are split up into five clusters; SU Oss is part of the wet cluster and is also heading this cluster, as a leading site in technology and volume (see appendix A). Other factories that produce similar sauces for Unilever are Sanguinetto in Italy (Bertolli) and a Co-packer Greencore (non-Unilever) in England. The raw materials are supplied to the SU and the internal customers of the SU are the local marketing and sales units (MSU’s). The MSU’s sell the products to the local external customers; mostly retail companies (see fig.1.2).

Fig. 1.2 European Supply Chain

Supply Chain

Market and Brand Development

Brand Activation and Sales European Supply

(12)

§1.2. Sourcing Unit Oss

Unilever Netherlands is divided into four major divisions; Foods, Ice cream and Frozen food, Home and Personal Care and Food solutions (see appendix B). There are five Sourcing Units; Baarn, Delft, Loosdrecht, Rotterdam and Oss producing product for several brands like: Becel, Bertolli, Blue Band, Calvé, Chicken Tonight, Conimex, Cup-a-soup, Knorr, Lipton Ice, Slimfast and of course Unox. SU Oss is part of the Unilever Food section and the plant is located in Oss since 1850. It has gone through a long history of mergers and investments which made the sourcing to what it is today. In 1937 the UNOX brand was created and twenty years later the first soups were launched. Late 1990 a new factory was built to produce sauces. Around 2001 when Unilever Bestfoods Netherlands was founded, SU Oss became an integral part of the Savory category. The plant is located in the center of Oss in an industrial area (see appendix C). Two large factories, with five hundred employees are located on the site.

SU Oss consists of two factories producing soups, sauces and meat. The conserved food (CFO) factory produces several kinds of meat products, including canned meat (mainly “frankfurters”), vacuum packed smoked sausages and the HEMA smoked sausages. The soup and sauce factory is divided into two production sections: the production of soups, i.e. Unox soups, and the production of hot sauces, i.e. Chicken Tonight. All fresh and frozen ingredients are delivered directly to the factories, the other ingredients and packaging materials are delivered to the central warehouse organization (CMO). All end products are distributed through this same CMO organization (see fig. 1.3).

Fig. 1.3 Logistics on plant site Delivery of fresh and

frozen raw materials

Central Warehouse (CMO) Meat factory (CFO) End product Intermediary product End product Packaging + ingredients

(13)

The three factories share a management team, lead by a works director. As whole team they are responsible for the total sourcing unit. The responsibility for the factories is divided between two factory managers; one for the CFO and one for the Soup- and Sauce factory jointly (see fig. 1.4)

Fig. 1.4 Organization Structure of SU Oss

§1.3. Markets

Over the last 5 years the SU Oss has increased in volume vast. In 2001 SU Oss produced 81.000 tons, increasing to 109.000 tons in 2006, thanks to an increase in efficiency and an expansion of the customer base. Customers are the Unilever MSU’s, organized per country. In 2001 the main customers were the Netherlands, Germany, Belgium and the United Kingdom. In the set up of the European Unilever and savory organization Oss has now exposure to customers in all European countries (see fig. 1.5), coordinated by a new Unilever Supply Chain Center (USCC), that will move to Switzerland later in 2006.

Fig. 1.5 Market expansions between 2001 and 2006 UK Germany Belgium Worksmanager Safety & Environment Factory Manager Soup & Sauce Factory Manager CFO Finance HRM Quality Assurance

(14)

The Netherlands is by far the biggest single customer, though the Dutch market has been under a lot of pressure because of the “supermarktoorlog” in 2005, where the largest Dutch supermarkets dropped their price dramatically, forcing the producers to lower their prices. The competition of B-brands and distributors own brands is also increasing. Unilever Foods has been monitoring itself against the main competitors. Figure 1.6 shows Unilever Foods has been better than the competition based on a competitor analysis of 2005. For 2005 Unilever foods shows the largest increase in value and in volume of all large fast moving consumer goods producers.

Fig. 1.6 Competitor analysis (Nielsen, 2005)

Unilever is the global leader in savory products with strong brands like Knorr and Unox, the savory market, however, is extremely competitive. Michael Porter has developed a model of five forces driving industry competition (Gibes, 1995: p. 5). The five competitive forces model is used for indicating Unilever’s competitive position in the savory market, and the strategies are crucial to remain competitive nowadays (see figure 1.7).

The threat of new competitors:

A significant barrier for entry is the high capital investment, necessary for the production of sauces. Next, brand recognition needs intensive marketing to reach brand loyalty. Of all companies in the Netherlands, Unilever invests the most in advertisement, about 220 million Euros. (Low threat, --)

VALUE 0.6 -2.7 -3.3 -7.1 -7.2 -10.4 -2.0 -1.4 -3.1 -4.8 -6.6 -7.9 -11.3 -15.0 -10.0 -5.0 0.0 VOLUME 2.2 1.3 -8.7 -9.2 -9.2 -6.1 1.9 -1.7 0.5 0.7 -4.9 -10.6 -9.0 -12.0 -8.0 -4.0 0.0 4.0 UNILEVER FOODS H.J. HEINZ NESTLE MASTERFOODS REMIA

UNILEVER FROZEN FOODS

UNILEVER HPC BEIERSDORF L’OREAL incl. GARNIER RECKITT BENCKISER HENKEL SCHWARZKOPF PROCTER & GAMBLE

(15)

Power of the buyers:

Due to the “supermarktoorlog” the power of the buyers, the retailers, is very high. There are many suppliers of different brands and private labels for sauces. Consequently, it is very easy for customers to switch between products. Customers are mainly large volume buyers, therefore, they are able to request very high service levels. (High power, ++)

Power of suppliers:

Unilever can choose from many suppliers for the supply of most ingredients. Therefore it can switch easily between several suppliers. Hence, the power of suppliers is not very large. (Low power, -)

Threat of substitute products:

It is always necessary to come up with new innovative products to remain competitive in this fast changing market. However, savoury products are not substituted very fast by other savoury products. (Neutral, 0)

Intensity of rivalry among industry competitors:

Unilever operates in a highly competitive and mature market (see fig. 1.7). There are many competitors, A and B brands, in this market. Unilever and its competitors use price competition, product innovations and intensive marketing activities to compete. This competition is strongly influenced by the existence of high exit costs, and mature markets where companies struggle to increase their market share. To sustain their competitive position in this market, Unilever has to distinguish it self from other companies. In today’s marketplace where technological innovation, outsourcing, e-business and global competition are prevalent, it is becoming increasingly important for Unilever to pursue Operational Excellence. Operational Excellence is critical to maintaining and advancing a company’s competitive edge. It results in world-class quality, productivity and customer service. In this pursuit, the implementation of modern manufacturing techniques is essential for a successful outcome. Furthermore Unilever strives to efficient production, a strategy called cost leadership, where production costs are controlled. Within SU Oss the emphasis lies on efficient production and reducing losses. Unilever uses Total Productive Maintenance (TPM) (see §2.4) as a help.

(16)

Fig. 1.7 Competitive forces in the savory market

§1.4. The Sauce factory

The Sauce factory is the smallest factory on site, with about 50 employees. Of all three factories, the Sauce factory is the most developed factory, with skilled operators working in self managing teams (see §1.5.). A large selection of sauces is produced in Oss for famous brands like Knorr, Chicken Tonight and Conimex (see appendix D). These products are distributed though all over the world with main customers in the United Kingdom, Hungary and the Netherlands. The factory runs five days a week, 24 hours a day. During the weekends, when there is no production, the maintenance and cleaning takes place. Order release is based on forecasts from the MSU. The MSU sends a forecast for short-term, long-term and an overview of the inventory in the warehouses. Based on this information the planning area of SU Oss provides the factory with a weekly operational planning, always four weeks in advance.

The chosen batch size is based on two aspects, firstly on the model for Economic Order Quantity and secondly on terms of delivery, agreed with the customer. In this agreement a minimum and a maximum delivery time is stated. Because of the large variety of product (156 different products), a production cycle is used, so not every product is made every single week and hence, batch sizes are larger. The planning area tries to group similar kind of products, with the same recipe or same jar, in the operational planning to reduce set-up times. Experience has shown that larger batches are most efficient, because then there are less changeovers and disturbances of the production-system.

(17)

§1.5. Organizational Structure

The soup and sauce production lines could be seen as one production area as they share equipment and both factories are run by the same factory manager (see fig. 1.8). The tactical and operational supervision in the Sauce Factory is executed by the Production Manager (PM). The Factory is also supported by two Process Improvement Managers (PIM) who back up the Production Manager with the tactical and operational supervision by taking care of an individual part of the primary process. Above all these PIM’s are responsible for improvements projects in the Sauce factory.

Fig. 1.8 Organization Structure of SU Oss

On the shop floor there are three types of operators; coordinating operators, sauce-operators and line-sauce-operators. They work in three teams (A, B and C) in a three shifts five days system, one team each shift. These three teams work as self-managing teams and consist of fourteen operators, where all three types of functions in a team are fulfilled. In each team there are two coordinating operators, who have an important role in the teams. Besides normal manufacturing-activities they are also responsible for additional tasks like improvement projects and the technical meeting every morning. The rest of the operators can be divided into two groups, sauce operators and line operators. Together they perform the normal manufacturing activities, where the sauce operators have special processing tasks on regulating the system and the control systems. As a team they are responsible for the production processes and the PM is their directly executive.

Factory Manager Soup Production Manager Self Managing Teams Process Improvement Managemers

(18)

§1.6. Main process flow

The production process can be divided roughly into two parts, a fabrication part and a packaging part. In the fabrication part the sauce is processed, filled and heat preserved. The packaging area labels, tray-packs and palletizes the products to its final state. Each box presents a processing station (see fig.1.9).

Fig. 1.9 Rough production process

Preparation of the ingredients

Before the actual processing can take place, all ingredients have to be weighted and set out so they can be added to the process. The ingredients that will be added in small portions less than five kilos have to be weighted at the spices-room. The special boxes with spices, precisely weighted per batch, and packaged per production run, are brought to the shop floor. The other ingredients can be divided into two categories. One part will be added through special “totes”, large, steely, movable containers, filled by the operator with spices, herbs, cheese and deep frozen vegetables and meat. The other ingredients like binding agent, water, peaches, tomato paste and dice are added directly through pipes, holding areas, filling tanks and fixed container-places.

Processing

Processing is the mixing and heating of several ingredients in large kettles. Processing consist of a number of steps, depending on the recipe of the product. These steps will be further clarified in the seventh section. The processing area is controlled from a special control room, where also the quality-inspection of the batches takes place.

Filling

In the filling section, jars are rinsed with hot water and subsequently filled with sauce. There are two types of fillers, a volume-filler and a vacuum-filler. After the jars are filled, they are transported to the capper that equips the jars with a cap. On this cap the inkjet printer prints a unique code, based on the product, production date and expiration date. Ingredients

Preparation

Processing Filling Heat

Preservation

(19)

Heat preservation

To maintain the quality of the sauces and prevent the sauce from rotting, the products are heat preserved. The goal of heat preservation is to extend the shelf life of the sauces. The sauce factory has two ways to preserve heat, sterilizing or pasteurizing. Sterilizing takes places in the retorts, a shared facility with the soup factory, for the pasteurizing a machine is embedded in the flow line.

Packaging

Finally, the products are labeled, tray-packed and palletized before going on transport to the distribution centers (DC’s) in Europe. End-products are stored first in the Central Warehouse Oss (CMO) for a short period, and from this stock they are transported to DC’s. From these DC’s, retail stores are supplied.

§1.7. Process Characteristics

The Sauce factory consists of only one production line where all different type of sauces are produced. There are some parallel machines, but they are never used in the same run. The line is a connected flow line for high volume production. There are only small buffers in the process, applied to intercept in case of short disturbances. When a large disturbance is perceived the whole line is brought to a halt.

The factory is set up with a product-oriented layout, so all sequential processes are performed in physically adjacent locations. This results in a clear flow of materials, where co-ordination should be easy. The products are pushed into the process, the traditional form of mass-production. Just like push-production, the sauce is produced in batches according to planning, moving downstream from station to station. Each production run consists of one ore more batches of the same product. Production must wait until all areas have completed their earlier run, change over and then are ready to process the new run. This waiting is quite unpredictable because there are so many different products produced in different batch sizes. Furthermore, quality is an important aspect in the Sauce factory. Samples of the sauces are inspected several times during the production processes. They are inspected on; pH-value, sweetness, structure, taste and color. Some of these measurements are qualitative, other quantitative, where the value has to be within a range, otherwise the product has to be corrected or rejected.

(20)

2. PROBLEM EXPLORATION

In a competitive environment, where Unilever operates, the performance of the production system is very important, as stated in the previous paragraph. The Sauce factory is yearly judged on several targets. This section discusses these targets in general and especially the targets Operational Efficiency and Performance Efficiency (PE), as these targets are used for measuring the performance and effectiveness of a production system. In pursuit of these last two targets, the Sauce factory is confronted with several problems. In the second paragraph these problems, discovered by management and operators, are expounded.

§2.1. Yearly targets: Operational Efficiency and Performance Efficiency

The output rate of the Sauce factory is very important as Unilever has several Sauce factories in Europe. There is internal competition as the total amount of sauce that has to be produced yearly for Unilever has to be divided into these factories. While the capacity of these factories is not enough to produce the yearly required volume, part of production is being outsourced to Greencore, a co-packer in England. So, a higher output is desirable, in order to produce a higher percentage of the annual volume in Oss. To accomplish this higher output, an efficient production system is important. The Sauce factory has specific yearly targets to measure the performance of the production system; Performance Efficiency (PE) and Operational Efficiency (OE). PE is an effectiveness factor that reflects the performance of a production system, the extent to whereas the production system has produced, according to its norm capacity, while the OE reflects the availability of a production system, the extent to whereas the production system has produced, related to the actual time the production was available for production. Every quarter of a year the factory reports its losses, based on the Unilever Pamco system, a Management Information System. The line operators record breakdowns and stoppages in QIS, a Quality Information System and the coordinating operator processes theses results into the Pamco system. These breakdowns are discussed each morning, during the technical meeting with the coordinating operator, the production manager and the mechanics. The calculation of the transfer prize is also partly based on the OE and PE. These transfer prizes are yearly determined, just as the targets for OE and PE. Other important targets for the Sauce factory are the Changeover time and the Zero Base

(21)

Waste and Safety. Changeover time is the number of minutes necessary to changeover the production system from one product to another. Products can differ in recipe, jar, cap or packaging material. Zero Based Waste is the percentage waste of material and packaging. Safety could be about product safety or safety of the employees, in this case the target is about the safety of the employees. Table 2.1 provides an overview of the performance on these targets from 2002 till 2005 and the target for 2006.

Target year 2002 year 2003 year 2004 year 2005 AC target 2006

PE (%) 74 % 74 % 80 % 80 % 86 %

OE (%) 67 % 67 % 71 % 72 % 76 %

Changeover time (min) 44 min. 45 min. 38 min. 35 min. 30 min.

ZBW (%) 3,1 % 4,3 % 3,9 % 2,7 % 2,5 %

Table 2.1 Overview results 2002-2005 & target 2006

The PE has to rise with six percent in comparison with 2005. The OE has to be raised also by four percent which would be a major improvement. Changeover time and the ZWB have to be reduced too. Remarkable in the table is that 2003 was a dramatic year according to these targets, because no improvements were made and some results even deteriorated. After 2003 many improvements were accomplished especially on ZBW, which was a major target for 2003 and 2004. For 2006 the main emphasis lies on improving the PE. But for all targets it is not clearly specified how these improvements will be accomplished, based on accurate figures. This research focuses on the performance of the production system and not on changeovers or waste, therefore the next paragraph will further expound the interpretations of the targets for OE and PE.

Within SU Oss the focus lies on the efficient use of production time as not all of the time available for production is used efficiently. The concept Total Productive Maintenance (TPM) is used for gaining more insight into the allocation of time and the improvement of the efficiency of the production system. Within Unilever a model, based on TPM, is used for analyzing the use of time and allocating time losses (see figure 2.1). This model contains three losses of time, Planned Non-Operational time, Routine Stoppages and Unexpected Stoppages. For the planned downtime and expected stoppages yearly targets are set up and these stoppages can of course never be reduced to zero. For expected stoppages many aimed improvement projects are set up to reduce these losses. For the unexpected stoppages it is also important to map them in order to set up improvement projects to reduce these losses to zero. But within the

(22)

Sauce factory not all unexpected stoppages are recorded. Based on this model, a loss analysis is expounded to seek for improvement opportunities. For 2005 the target for PE was set on 86%, but the obtained PE for that year was only 80%.

Fig. 2.1 Overview of use of time

§2.2. Recent issues in the Sauce factory

OE and PE are important targets for the Sauce factory, but in previous years the Sauce factory was not able to meet these targets. There are several issues in the Sauce factory that affect the effective use of time (PE) and thus also the OE and net output rate. Some issues are brought by the management, operators and mechanics, other issues were noticed in the field.

§2.2.1. Problems regarding Production Rates

An important issue brought up is that it is unclear at which production rate the system produces ideally. The production rate of most machines rate can be adjusted by the

Planned Loading Time L O P E Opera- tional Time Production Time Effective Time Planned Non Operatio-nal Time Routine Production Stoppages Unex-pected Stoppages Planned - Trials Downtime: - Buildings - Maintenance Expected - Start-ups Stoppages: - Changeovers

Unexpected - Major Breakdowns Stoppages: - Minor stoppages

(23)

operators, or is controlled by automatic regulations. There are prescribed rates (from 2001) for the filling area, but these rates have not been revised since introduction. These rates are probably outdated and the theoretical background of these rates is unclear. As a result, each operator is forced to adjust the production rates in his own point of view. The operators do not think the current prescribed rates are credible, which leads to insecurity and confusion among the operators. The lack of appropriate communication between areas makes fine-tuning almost impossible. Not only is it hard to set up an appropriate rate, it is also important that all areas are united in one continuous flow. At this moment the flow of the areas is not being adjusted in an appropriated way. It is not easy to adjust production rates, because of the heat preservation, this stage delay in the process of over 45 minutes. There is not enough insight within SU Oss, on how to react to variances in production rate due to unexpected stoppages. Above all, more insight is needed into bottlenecks in the line. What are the bottlenecks and what is the effect on the system? The lack of this information impedes a steady and continuous flow. Several small buffers are placed in the line, but operators tend not to use these buffers. All machines are coupled in one continuous flow line. When a machine downstream is producing at a lower rate then a machine upstream and there is no buffer to intercept this difference in production rate, products get jammed. When this jamming takes over 30 minutes, and the sauce has not been heat preserved yet, the sauce has to be drained because the quality is not guaranteed.

Problems regarding Production Rates

Outdated Production rates.

Lack of theoretical background of production rates.

Confusion with the operators, because of implausible prescribed rates. Bad adjustments of production rates between areas.

Buffers are not used in the right way.

Lack of insight on how to deal with variances in production rates. Jamming and Starvation occur frequently.

Lack of insight into bottlenecks.

(24)

§2.2.2. Problem regarding Norm Rates

The lack of plausible production rates also affects the Sauce factory on another level, while norm rates are used as a standard for financial settlement and planning. But these norm rates are not reliable because they are also outdated and have not been adjusted for a long time. Resulting, the planning department is not able to make an accurate planning, especially as there are shared resources and adequate planning based on reliable norm rates is essential. Besides the planning department, the financial department also uses the norm rates for determining the OE and PE. So the target stated for 2006 is based on unreliable information. Like last year there is a fair chance that this leads to a gap between the obtained PE/OE and the target for 2006. Customers pay the transfer prices and not the actual production costs. Therefore a large gap between the actual capacity and an outdated norm capacity could also result in a gap between income and expenses.

Problems regarding Norm Rates

Planning problems

Non-transparent financial settlement; Gap between income and expenses Table 2.3 Summary of problems regarding Norm Rates

§2.2.3. Problems regarding Loss of Performance Efficiency

The quarterly loss reports show a big problem; the large amount of unexpected stoppages. About twenty percent of the time that can be used for production is lost because of these unexpected stoppages, while these stoppages should be reduced to zero. Idling and minor stoppages occur frequently. The machine actually stops producing if there are no products available to work on (starvation or idling) or when the machine itself stops less than ten minutes (minor stoppage / blocking / jamming). A minor stoppage could be the result of for example a small defect. These short stoppages reduce the net production rate and the Performance Efficiency of the system, but are barely recorded and mapped. Only the big breakdowns are mapped in the Pamco-system. Of most of the unexpected stoppages it remains unclear through what they are caused and how long and how often they occur. Therefore it is nearly impossible to take appropriate actions and set-up improvement-projects. Even harder to record are the losses as a result of reduced speed. When a machine is producing on a rate lower then

(25)

its ideal rate, based on the machine capacity, performance loss occurs. The operators often also cause reduced speed here themselves. Operators tend to reduce the rate of machines so that less disturbance occurs. Most of the time the production rate is tuned lower then the prescribed rate because the operators want to prevent big breakdowns, the operators expect more big breakdowns at a higher production rate. So, unexpected stoppages are hard to focus on, because only big breakdowns are recorded by the operator and filed into Pamco. As a result only part of the time losses could be allocated.

Problems regarding Loss of Performance Efficiency

20 percent unexpected stoppages.

Operators tend to reduce the rate to prevent stoppages. Lack of aimed improvement projects.

Suspicion of large loss through reduced speed. Idling and minor stoppages.

(26)

3. RESEARCH DESIGN

In section two the first motive for this research was phrased by giving an overview of recent problems (see paragraph 2.2.). Subsequently, this section starts with an analysis of these problems, where mutual dependencies between problems, as well as the problem holders are discussed. Deducted from that analysis the problem statement is expounded, starting point of the research. This research design furthermore describes the line of actions.

§3.1. Problem-analysis

In section two several problems were elaborated (see Appendix E.). It is essential to review these problems and to see whether and how they are related, in order to identify the central functional problem. A functional problem is always visual in the output of a system; in this case it is the Sauce factory. The problems within the Sauce factory can be divided into three parts; problems regarding production rates, norm rates and the loss of performance. But these problems cannot be seen as individual problems, because they are related. The three problem areas and their mutual dependencies are described and reflected in a picture, see figure 3.1 at the end of this paragraph. Within SU Oss there is not enough insight on how to deal with production rates in the manufacturing area. There are no specifications of the capabilities of the production processes and there is no insight into bottlenecks. This lack of insight has a major impact on the output of the Sauce factory; manufacturing problems on the shop floor on one hand and unreliable norm rates on the other hand.

Firstly problems on the shop floor exist because there are no reliable prescribed rates. The operators do not know how to control the production system by adjusting operating rates of different machines and areas. Adapting to variation in production rate of different stages of the production process plays an important role. The operators have no insight into these variances. Furthermore, there is no insight into what influences the production rate that is tuned; there is no clearance of capacity and bottlenecks. This leads to idling, minor stoppages, reduced speed and big breakdowns. There is also quality loss in case of big breakdowns because part of the product has to be pumped away because the quality is no longer guaranteed after these big breakdowns. These are all forms of

(27)

performance loss. These losses are not only problems of the operators, but also of the management. The idling and minor stoppages are not registered, so aimed improvement projects are hard to set-up. The lack of reliable norm rates leads to unreliable planning. This could be stressful for the operators if the calculations are too high and could lead to deliver problems. On the other hand when the norm rates are to low, less is produced than possible because not all capacity is used.

The norm rates are also used for performance measurement. If the way of measurement and the current norm rates are not adequate, this leads to inaccurate information regarding performance. This information is important for the financial department. The norm rates are based on outdated information and lead to unreliable targets for OE en PE, which could form a large problem for the factory management, because the performance of the factory is calculated on these figures. The target for Performance Efficiency for 2006 is 86 percent. Every week the performance is measured and published throughout SU Oss. For 2005 the factory was not able to reach the target of a PE of 86 percent and in the first weeks of 2006 this trend was continuing. Therefore, problems described have a causal relationship, because the gap between target and performance might be partly the result of unreliable norm rates. Not all losses registered, unreliable norm rates and the loss of performance efficiency lead to a gap between the targets OE / PE and the actual obtained OE / PE. This leads to other functional problems besides deliver problems, namely incorrect transfer pricing, profit loss and loss of bonuses for management and operators. The incorrect transfer pricing is a problem for customers as well as management, because there is a gap between income and expenses. This diminishes the profit and the shareholder value.

So the organization needs information on these main subjects; creating more insight into the production system, production rates and bottlenecks, formulating adequate norm rates and using the production system more efficient by improving performance efficiency. Central in this problem area and base for these functional problems is the lack information on production rates. Therefore this research will focus first on how to deal with production rates and through what they are influenced. An overview of the relative context of the problems, as explained above could offer more insight into the problem area (see figure 3.1).

(28)

Fig. 3.1 Relative context of problems

Lack of insight in production rates &

bottlenecks

Unreliable norm rates Lack of feasible prescribed rates

Idling & m inor stoppages

Loss of bonus for m anagem ent & operators

Not recording all losses

gap between target O E / PE and real O E / PE

Production delays

Gap between incom e and

expenses Profit Loss

Jam m ing Big breakdowns

Reduced Speed

Loss of perform ance efficiency Incorrect adjustm ent of dif. production rates of

areas Lack of credility of prescribed rates by operators Unreliable schedeling Unreliable target O E PE Product loss Unexplained losses Lack of aim ed im provem ent projects

Deliver Problem s

Problem s regading Production rates

Funtional problem s

Central Problem Problem s regarding Loss of

Perform ance Efficiency Problem s regarding Norm Rates

(29)

§3.2. Problem Statement

As a result from the problem-analysis a research is setup on production rates, norm rates and Performance Efficiency. In literature a lot of information can be found on Performance Efficiency and production rates in general, but the specific character of the Sauce factory, a food processing industry, may affect this research. Therefore a literature review is done, also specifically on Food Processing Industries, in order to gain more insight into factors influencing Performance Efficiency. Combined with special characteristics for this case, at SU Oss a model will be created with factors influencing the Performance Efficiency and production rate. Based on this model the current situation at the Sauce factory will be analyzed.

Research objective:

The research objective is deducted from the problem analysis and the model above.

The objective of this research is to gain more insight into the capabilities, the bottlenecks and the performance of the production system of Unilever’s Sauce factory in Oss, in order to improve the net production rate.

Preliminary Diagnostic research question:

A diagnostic sub question can offer us more insight into the knowledge objective of the research.

Through which factors is the net production rate influenced in general and specifically at the Unilever SU Oss Sauce factory?

A way to structure a research is to use sub questions to streamline the research process, therefore five sub questions are phrased below.

(30)

Diagnostic sub questions:

Through which factors are production rates influenced in general and specifically for the food processing industry? (Literature review)

How are these factors influenced by specific characteristics of the Sauce factory?

Through which factors is the net output rate, the performance of the production system, influenced in the Sauce factory?

§3.3. Research Approach

Final product

Final product of this research is a report with redesign of the feasible production rates and the norm rates and insight on how to deal with production rates. Besides this, this research provides the management of the Sauce factory with recommendations on how to optimize the production system and increase Performance Efficiency.

For this research some research boundaries can be defined:

• The research will focus on the Sauce factory of Unilever SU Oss.

• The research will only focus on the area’s processing, filling, heat preservation and packaging.

• Based on the allocation of the use of time, this research will only focus on the Performance Efficiency and the Operational efficiency, though changeovers and start ups are examined.

• The research has to be finished by End-August.

Project composition

The project consists of three stages, directly deducted from a research model of De Leeuw (Leeuw, de, 2001); diagnosis, design and change. This research is mainly limited to the first two stages, the change phase, or implementation of the (re)design, will be accomplished by the Sauce factory. The recommendations in the final section will be the point of departion for this stage. Figure 3.2 describes the line of actions.

(31)

Fig. 3.2 Overview of research

Research type:

This research consists of two parts; a literature review which has an explorative character and a case study at Unilever.

P r e lim in a r y

r e s e a r c h

D e s ig n

D ia g n o s is

1 . C o m p a n y D e s c r ip t io n 2 . P r o b le m E x p lo r a t io n 3 . P r o b le m A n a ly s is & R e s e a r c h D e s ig n 4 . T h e o r e t ic a l F r a m e w o r k 8 . B o t t le n e c k A n a la y s is 9 . A n a ly s is o f n o r m r a t e s a n d P e r f o r m a n c e M e a s u r e m e n t 1 0 . P e r f o r m a n c e lo s s e s 1 1 . C o n c lu s io n s a n d d e s ig n S e t u p 1 3 . N e w n o r m s a n d I m p r o v e d p e r f o r m a n c e m e a s u r e m e n t 1 4 . R e d u c in g P e r f o r m a n c e lo s s e s C o n c lu s io n s a n d R e c o m m e n d a t io n s 5 . C o n c e p t u a l F r a m e w o r k 7 . P r im a r y P r o c e s s 6 . M e t h o d o lo g y 1 2 . B o t t le n e c k I m p r o v e m e n t s

(32)

Definitions

For unequivocality a few keywords of this report are defined below, a complete overview can be found at the end of this report.

• Net Production Rate:

Net production rate is the actual rate at which the production system produces, calculated by the number of jars produced in the run divided by the effective production time of a run.

• Prescribed (production) rate:

This rate is the defined optimum rate for a particular product. These production rates are used on the shop floor as instructions for the operators.

• Bottleneck:

A bottleneck here is a (part of a) workstation of a process that causes the entire process to slow down or stop.

• Norm rate:

The defined norm production rates for a particular product, these rates are used for planning and targets.

• Performance Efficiency (PE):

PE is an effectiveness factor that reflects the performance of a production system, the extent to whereas the production system has produced, according to its norm capacity.

• Operational Efficiency (OE):

OE is an effectiveness factor that reflects the availability of a production system, the extent to whereas the production system has produced, related to the actual time the production was available for production.

(33)

4. THEORETICAL FRAMEWORK

This section consists of a review of literature on production rates, performance measurement and Performance Efficiency. Two types of production are distinguished in the previous sections; theoretical (prescribed/ norm rate) and performance (net production or output rate). The central problem is the lack of prescribed rates and reliable norm rates is. Without reliable norm- / prescribed rates it is not possible to check the performance of the production system, because the performance of the system is directly depending on the existence of suitable norms. Therefore the primary focus of research is determining feasible net production rate and norm rates. Comparing the norm rate with the net production rate is the second step of the research. This research also examines factors influencing this gap, the feasible production rate and the performance (see figure 4.1). In literature, many concepts are used for analyzing production rates. Several of the latest concepts into production rates are reviewed in order to reflect trends in these topics; production rates in general, capacity, bottleneck, flow production, other limitations, performance measurement, Performance Efficiency and production control. This literature review leads to a framework of theories which will be reflected in a conceptual model of factors influencing production rate in the next section.

Fig. 4.1 Overview of production rates

§4.1. Flow production

Process types influence the control and the rate of production. Bertrand, Wortman and Wijngaard (1998: p. 144) distinguish several production situations. As stated in the first section, the Sauce factory production system is a form of flow production. Flow production is one of the typical situations, characterized by workstations that are available at the same time, roughly equal operating times and all products having the

Prescribed production rate

Net production rate (performance)

(34)

same routing. Flow process-manufacturing is a typical flow process, where the difference between factory logistics and department-control fades. A relatively stable product-assortment and a relative high demand per time unit characterize flow process manufacturing. Further there is a relatively high production rate per product type. Due to the integration of the different stages of the manufacturing process each production line often focuses on one product type at the same time, which simplifies production control (Bertrand et al., 1998: p. 493). Like the assembly line, a continuous flow process has a fixed pace and fixed sequence of activities. Rather than being processed in discrete steps, the product is processed in a continuous flow; its quantity tends to be measured in weight or volume. In a continuous flow production system the production rate of the total system depends on the capacity of the system, because all the stages are connected and queues do not exist in the ideal flow situation (Bertrand et al, 1998: p. 144). Therefore, there is a total absence of waiting time and every station has the same tact time (Bertrand et al 1998: p. 145) or operating time, so that there is a regular flow from station to station. When an impediment occurs in the pure flow-production buffers are needed (see §4.7). Bertrand, Wortman and Wijngaard (1998: p.148) write about factors influencing this pure production. There are seven reasons why a pure flow-production does not succeed; unequal flow-production times for different workstations, variances in production time per station, different products, workstations sensible for disturbances, appropriate technology for production is not available for desired tact time, learning curve and variance in routings (Bertrand et al., 1998: p. 147).

§4.2. Production rates

A major challenge for industrial engineers is to design and operate production lines efficiently. Chen and Askin (2006: p.6) refer to production rate as a major key of interest. Production rate can be calculated by dividing the available time by the number of products that is produced. There are several formulas for calculating or estimating the production rate, like for synchronous and asynchronous line, two or more serial stations. But most of these formulas are based on specific situation that do not correspond with the production situation in the Sauce factory. Most formulas do not consider controllable machine speeds and are thus not applicable. Strusevich (Hall, 1996: 518) does consider controllable production rates, but his formula is only applicable in case of a no wait two machine situation. Tan (1999: p. 100) also considers production rate as a first order

(35)

performance measure. According to Tan, a zero buffer line must be analyzed to determine the feasible production rate. But this analysis is not applicable to the Sauce factory as there a few buffers placed in the line. Liberopoulos and Tsahouros (2002: p. 65) provide four ways for increasing the effective production rate of a transfer line: (1) by increasing the capacity of the workstations, starting with the slowest, (2) by reducing the frequency failures, (3) by reducing the duration of failures and (4) by reducing the impact of failures. A second order performance measure discussed in this paper is the variability or variation of net production rate. Gershwin (1992) presents a method to determine the variance of the output in a given time period from a single station. This method is based on deriving the different equations for the probability of producing n parts at a given time and then solving these equations by using some boundary equations. But in this case there are too many boundaries and workstation, so this is not a proper formula for the situation in the Sauce factory.

§4.4. Production capacity and bottlenecks

The capacity of the process determines its maximum output rate, measured in units produced per unit of time. Capacity is initially restricted by the maximum capacity of a stage or machine. The isolated capacity of a machine means that this machine can never be blocked or starved (Helber, 1998: p. 29). The process capacity is determined by the slowest series task in the process; that is, having the slowest throughput rate or longest cycle time. This slowest task is known as the bottleneck (Gibson, Greenhalgh and Kerr, 1995: p. 69). Identification of the bottleneck is a critical aspect of process analysis since it not only determines the process capacity, but also provides the opportunity to increase that capacity. The term bottleneck is derived from the narrow part of a bottle used to slow down the flow of liquid. However, here the bottleneck metaphor describes a link in a process that tends to slow down the entire process. Saving time in the bottleneck activity saves time for the entire process. Saving time in a non-bottleneck activity does not help the process since the net output rate is limited by the bottleneck. It is only when the bottleneck is eliminated that another activity will become the new bottleneck and presents a new opportunity to improve the process. Time is lost at the bottleneck when idling occurs. Idle time is time when no activity is being performed, for as a result from starving or blocking. Starvation occurs when a downstream activity is idle with no inputs to process because of upstream delays. Blocking occurs when an

(36)

activity becomes idle because the next downstream activity is not ready to take it (Gibson, 1995: p. 70). If the next slowest task is much faster than the bottleneck, then the bottleneck is having a major impact on the process capacity. If the next slowest task is only slightly faster than the bottleneck, then increasing the throughput of the bottleneck will have a limited impact on the process capacity.

An important concept on these grounds is the Optimized Production Technology (OPT) (Slack, 1998: p. 539), developed by Eliyahu Goldratt (the Goal, 1984). The core of planning production is dealing with bottlenecks, an hour lost at a bottleneck is an hour lost for the entire system. This is one of the nine basic principles of the OPT (Gibson, P. 1995: p.149). Another central theory regarding capacity is the Theory of Constraints (TOC). The TOC is a management philosophy developed in the early 1980s, also by Goldratt. A constraint is anything that limits a system’s performance relative to its goal. Constraints can be physical/logistical, managerial/procedural or behavioral/psychological and there is also a difference between internal and external constraints (Ruhl, 1996: p.49). Therefore managing constraints is the main goal. TOC provides a five-step process for managing constraints, which can be used as a part of an effort of continuous improvement. Remove the constraint, the theory goes, and you increase the rate of production and boost the bottom line. First step in this theory is to identify all systems constraints, also called bottlenecks. The other four steps of TOC are; exploiting the systems constraints, subordinate everything else to do so, elevate the systems constraints and repeat al five steps if a constraint is broken (Miller, 2000: p.49-53).

§4.5. Characteristics of the food processing industry

In 2001 22 percent of the total turnover of the Dutch industry is provided by the food industry (www.cbs.nl). A thorough look at the production management research literature reveals that a majority of it has been for industries handling discrete units, fabricated and / or assembled during manufacturing. The food processing industry is strongly influenced by several special characteristics. These characteristics are not only different from discrete industries, but also different from several other process industries, like oil or other chemical industries. In the previous paragraph is stated that the capacity of the processes is restricted by constraints. The special characteristics of these industries might affect the constraints limiting the capacity. and request a different approach of production control. Gaalman, Van Donk and van Wezel (2005: p. 7) divide several

Gaining insight into capacities, bottlenecks, performance measurement and Performance Efficiency at Unilever’s Sauce factory in Oss