Trends in Food Innovation

8/15/2019 Trends in Food Innovation

1/20

Trends in Food Innovation: More Than Secret Recipes

Speed-to-market and innovation are often associated with technology-focused

industries like high tech or automotive. ut these !"Is are also important to one

of the oldest industries in the world: the food industry.

The food industry is as old as civili#ation$ and many of its process operations are

thousands of years old% such as &rewing 'developed in Sumeria and a&ylon( and

&aking 'developed in )gypt ca. *+++ ,(. The modern food manufacturing

industry evolved during and after the Industrial Revolution and today the food

industry is going through another important change process. Several trends

impose new challenges on food manufacturers: glo&al food regulations%

demanding customers who ask for sustaina&le products% the trend of functional

foods% new reuirements for la&eling and tracea&ility% and much more.

et/s look at some of these trends:

,utting food waste is the top trend in 0+12. 34aste not want not5 re6ected

manufacturers/ e7orts to reduce food loss and waste during the production

process. 8ue to poor practices in harvesting% manufacturing% storage and

transportation% as well as market and consumer wastage% it is estimated that 9+

;+< 'or 1.00 &illion tons( of all food produced never reaches a human stomach.

The => Food and ?griculture @rgani#ation 'F?@( works with the international

engineering community to ensure governments of developed nations put in place

programs that transfer engineering knowledge% design know-how% and suita&le

technology to newly developing countries.

Food safety and tracea&ility are demanded &y customers. In ,hina% a technology

Arm recently launched a chopstick that tests your food and tells you if it is safe

to eat. It is not clear whether the smart chopsticks will go into commercial

production. The company had made only a limited run of prototypes% &ut there is

a huge interest according to discussions in social media forums in ,hina. ut

food safety should not start with the end consumer% it should start right at the

&eginning during the innovation process.

8id you know that 98 printing has already arrived in the food industryB Read this

interesting &log in which my colleague Richard Cowells talks a&out Cershey/s

partnership with a 98 printing company and arilla/s plans to position 98printers for pasta in restaurants. Imagine going to a lovely Italian trattoria that

prints your 3linguine5 Dust the way you want them. 4ell% does not sound very

romantic% &ut deAnitely interesting.

ong story short% food recipes and food processing instructions are no longer

kept in Erandmas 3secret cook&ook5% they are managed and developed in

compleG% integrated IT systems.

For RH8 eGperts in process industries% it is daily &usiness to connect operating

silos% streamline the ramp-up to production% manage compleG product data%

perform compliance checks% and much more. Their goal is to improve the valueof their &rand &y launching new% successful products. "roduct ifecycle

http://www.foodmanufacture.co.uk/Ingredients/Top-10-food-and-beverage-trends-for-2014http://www.theguardian.com/world/2014/sep/04/baidu-china-search-engine-smart-chopsticks-food-safetyhttp://www.forbes.com/sites/sap/2014/01/24/chew-on-this-3d-food-printing-could-soon-be-a-reality/http://www.theguardian.com/world/2014/sep/04/baidu-china-search-engine-smart-chopsticks-food-safetyhttp://www.forbes.com/sites/sap/2014/01/24/chew-on-this-3d-food-printing-could-soon-be-a-reality/http://www.foodmanufacture.co.uk/Ingredients/Top-10-food-and-beverage-trends-for-2014


2/20

Management software from S?" helps food manufacturers to develop formulas%

manage the reuse of ingredients% run analytics% and design the packaging in

one environment.

Rich "roducts ,orporation% a family-owned fro#en food manufacturer and

solutions provider &ased in u7alo% >ew ork% recently shared their eGperienceswith S?"Insider maga#ine. 4ith the recipe development functionality in S?" "M%

the Anished product and all its components 'recipes% formulas% ingredients%

packaging% speciAcations% nutritional% and la&eling information( are linked

together. That integrated data is one in the same that is used for managing the

supply chain. Information feeds all the way down to the &ill of material on the

shop 6oor% providing a tighter integration &etween RH8 and "roduction. The

company eGpects some signiAcant &eneAts from the implementation in the near

future: reduction in cycle time% increased revenue from new product

development% faster product delivery% valua&le real-time information to users%

and work6ow enhancements that improve user productivity and mitigatecompliance issues.

Tate H yle% a glo&al food ingredient producer% a glo&al provider of ingredients

and solutions to the food and &everage industries% implemented S?" "roduct

ifecycle Management to standardi#e their innovation processes for new

ingredients% to have a central access point for all recipes and to link the RH8

team with production. 4atch the video.

4ith all the new technologies that help food manufacturers develop sustaina&le%

tracea&le% high-uality products within shorter time% the food industry should &e

a&le to focus on its core competence: delicious recipes like the ones that can&e found in my Erandmas secret cook&ook.

- See more at: http:JJwww.news-sap.comJtrends-food-innovation-secret-

recipesJKsthash.MLG!F.dpuf

Business looks pretty sweet for Tate H yle. New global opportunities abound for the London-based company,

which makes its bread and butter supplying ingredients to the food and beverage industries; soon it could also

be providing the sweetener used on your dining room ta&le.

http://www.sap.com/bin/sapcom/en_us/downloadasset.2014-10-oct-01-11.delivering-products-to-customers-faster-with-sap-product-lifecycle-management-pdf.bypassReg.htmlhttp://scn.sap.com/community/business-trends/blog/2014/06/19/sap-s-secret-sauce-for-transforming-a-global-businesshttp://scn.sap.com/community/business-trends/blog/2014/06/19/sap-s-secret-sauce-for-transforming-a-global-businesshttps://www.youtube.com/watch?v=6ilyudhtsakhttp://www.news-sap.com/trends-food-innovation-secret-recipes/#sthash.BM9xVKFV.dpufhttp://www.news-sap.com/trends-food-innovation-secret-recipes/#sthash.BM9xVKFV.dpufhttp://www.tateandlyle.com/http://www.tateandlyle.com/http://www.tateandlyle.com/http://www.foodbusinessnews.net/articles/news_home/Financial-Performance/2014/05/Tate__Lyle_free_to_explore_tab.aspx?ID=%7B65F9F2ED-C7B7-4A45-9BDC-88EAA6994C67%7D&cck=1http://www.foodbusinessnews.net/articles/news_home/Financial-Performance/2014/05/Tate__Lyle_free_to_explore_tab.aspx?ID=%7B65F9F2ED-C7B7-4A45-9BDC-88EAA6994C67%7D&cck=1http://www.foodbusinessnews.net/articles/news_home/Financial-Performance/2014/05/Tate__Lyle_free_to_explore_tab.aspx?ID=%7B65F9F2ED-C7B7-4A45-9BDC-88EAA6994C67%7D&cck=1http://www.sap.com/bin/sapcom/en_us/downloadasset.2014-10-oct-01-11.delivering-products-to-customers-faster-with-sap-product-lifecycle-management-pdf.bypassReg.htmlhttp://scn.sap.com/community/business-trends/blog/2014/06/19/sap-s-secret-sauce-for-transforming-a-global-businesshttp://scn.sap.com/community/business-trends/blog/2014/06/19/sap-s-secret-sauce-for-transforming-a-global-businesshttps://www.youtube.com/watch?v=6ilyudhtsakhttp://www.news-sap.com/trends-food-innovation-secret-recipes/#sthash.BM9xVKFV.dpufhttp://www.news-sap.com/trends-food-innovation-secret-recipes/#sthash.BM9xVKFV.dpufhttp://www.tateandlyle.com/http://www.foodbusinessnews.net/articles/news_home/Financial-Performance/2014/05/Tate__Lyle_free_to_explore_tab.aspx?ID=%7B65F9F2ED-C7B7-4A45-9BDC-88EAA6994C67%7D&cck=1


3/20

SAP MaxAttention has helped Tate & Lyle go to

market faster with ingredients for new recipes,

as well as operate in dierent regulatory

enironments across the glo!e"Sweetening the pot, ate ! Lyle recently announced a shareholder dividend for this summer, and at least

one analyst firm has raised its price target on the company/s stock.

But it"s not all cupcakes and candycanes for ate ! Lyle, which is still undergoing a transformation put

forward by the company"s #$% in &'('. )aking ate ! Lyle stronger and more diverse re*uires changing the

company"s processes and increasing its efficiency.

+e want to develop ingredients for new recipes and bring them to market faster ... and we need to operate in

global markets with different regulatory re*uirements, ate ! Lyle #% Steve Byers told S?" in a company

video. +e also wanted to take advantage of industry best practices.

ate ! Lyle is transforming with the help of S?" MaG?ttention, an solution that enables the company run

more smoothly while remaining agile enough to work across the globe. Shortening the design debate about best

practices was )a/0ttention"s +secret sauce, according to Byers.

+ had to decide how to support two ma1or business units in a common way across the globe, Byers said. +hat

established the strategy of a single global platform based upon S02.

)a/0ttention offers long-term high-level support to organi3ations that partner with S02. he solution is

especially useful for lengthy activities -- such as ate ! Lyle"s transformation -- because it4

• ,olla&orates with onsite support teams% enhancing technical eGpertise and account

management skills

• ,o-innovates with partners% deriving additional value from S?"/s maintenance programs

• @ptimi#es performance and simpliAes maintenance of S?" solutions

+e sped through the build process, meeting our deliverables on time, Byers said. +0nd the biggest reason we

were able to do that is because the design was solid.

ate ! Lyle baked into its pro1ect an nnovation #ontrol #enter, which reviewed solution re*uests to minimi3e

custom resolutions -- optimi3ing best practices from the start. 5edicating company resources to work with S02

was crucial to the value ate ! Lyle continues to derive from )a/0ttention, according to Byers.

http://www.mideasttime.com/tate-lyle-plc-price-target-raised-to-gbx-730-at-numis-securities-ltd-tate/155806/http://www.mideasttime.com/tate-lyle-plc-price-target-raised-to-gbx-730-at-numis-securities-ltd-tate/155806/http://herald-review.com/business/local/tate-lyle-continues-transformation/article_4eb541ae-65db-5481-aec0-5ac8afd5040c.htmlhttp://www.youtube.com/watch?v=WJmBBH7nbQIhttp://www.youtube.com/watch?v=WJmBBH7nbQIhttp://www.youtube.com/watch?v=WJmBBH7nbQIhttp://www.youtube.com/watch?v=WJmBBH7nbQIhttp://www.sap.com/services-support/support/plans/info/maxattention/index.htmlhttp://www.sap.com/services-support/support/plans/info/maxattention/index.htmlhttp://www.mideasttime.com/tate-lyle-plc-price-target-raised-to-gbx-730-at-numis-securities-ltd-tate/155806/http://www.mideasttime.com/tate-lyle-plc-price-target-raised-to-gbx-730-at-numis-securities-ltd-tate/155806/http://herald-review.com/business/local/tate-lyle-continues-transformation/article_4eb541ae-65db-5481-aec0-5ac8afd5040c.htmlhttp://www.youtube.com/watch?v=WJmBBH7nbQIhttp://www.youtube.com/watch?v=WJmBBH7nbQIhttp://www.sap.com/services-support/support/plans/info/maxattention/index.html


4/20

+he transformation remains firmly on track, ate ! Lyle ,)@ Naved ?hmed told industry analysts last

month.

0nd as the company continues moving into new markets, such as tabletop sweeteners, its prospects still look as

sweet as ever.

Get maximum tailored support for the long haul –

with SAP MaxAttentionake advantage of the highest level of support and enter into a long-term engagement with S02. his

customi3ed, collaborative premium support partnership complements the S02 $nterprise Support offering and is

ideal for environments whose activities span multiple years and pro1ects.

• ork with your onsite support team to receive e/tensive technical e/pertise and overall in-depthtechnical account management

• 0ccess ongoing co-innovation and co-value reali3ation and derive additional value from S02"s

maintenance programs

• %ptimi3e the performance and maintainability of your S02 solutions

Shell and SAP Co-Innovate With the SAP HAA Platform

September 6', &'(7 8 S02 - 5atabase echnology

S02 N$SB9$ - S02 S$ :N9S$4 S02 today announced that Shell and S02 have been collaborating to

develop a new well and reservoir facility management :0N0?. Shell

recogni3es the value of implementing S02 >0N0 as a single in-memory computing platform, which can help

accelerate analytics, business processes and sentiment data processing.

+#ustomers such as Shell can capitali3e on the opportunities that have been opened up through S02 >0N0,

said Steve Lucas, president, 2latform Solutions, S02. +By co-innovating with customers, we are seeing the direct

impact S02 >0N0 has on customer landscapes.

!uilding a Customi"ed Solution powered #$ SAP HAA

Shell is an innovation-driven global group of energy and petrochemical companies, which is active in more than

@' countries worldwide. n order to enhance performance of e/isting applications, Shell and S02 plan to

collaborate to create the ne/t-generation toolkit for well and reservoir reviews. he envisioned solution shall

deliver tools and best practices empowering users to build a professional understanding of their assets and

identifying opportunities in the operate phase. his aims to result in an increased early production and improved

optimi3ation of resources.

he planned


5/20

the first step toward a renovated smart solutions platform in Shell and an e/cellent e/ample for leveraging the

S02 >0N0 platform capabilities in a non-traditional S02 domain.

Cow )R" implementation &uilds Os world famous Supply ,hain:

Virtual Integration: When ERP fts the Dell's Direct

model:

The introduction of enterprise resource planning ')R"( software improves the

coordination &etween Arms. efore )R"% the each function in value chain had

separate organi#ation with separate information system. )ach function

performed its own tasks thus not glo&ally optimi#ing the whole value chain. )R"

&uilds the Pelectronic nervous systemP to links all units together and increases

overall productivity.


6/20

In some cases% Arms found that they could eliminate most inventories &y shifting

to faster &ut more eGpensive transportation alternatives 'e.g. air cargo( that

replenished supply Dust in time. Simply put% )R" allowed information to replace

inventory.

The emergence of the Internet facilitated more and more information sharing

&etween Arms% eGtending the &eneAts of )R" from the value chain of an

individual Arm to the entire value system of Arms and their suppliers and

customers. )R" can &e a vital component in controlling compleG supply chains

and in the fast developing world of e-&usiness and 0 electronic eGchanges.

8ell ,omputerOs success in reducing ineQciencies esta&lishes it as a model for

many other companies.

The Dell Model

8ellOs success is &ased on reali#ing the strategic power of the supply chain. The

core of the 8ell model is to deal directly with and sell directly to the customer%

and &uild products to order. 8ell collapses the value chain and eliminates two

signiAcant cost components: the retailerOs mark-up and the costs and risks

associated with carrying large inventories of Anished goods.

TeGas-&ased 8ell is the worldOs largest personal computer maker. Founded in themid-1L*+s &y a university student% Michael 8ell% the company leads the sector

with annual growth rates of 9+ to 2+ per cent. 8ell has achieved its success in

large part due to its highly eQcient value chain integration approach% supported

&y )R" and - more recently - &y the Internet. 8ell produces custom-made

computers PDust in timeP for orders received directly from the customer via

telephone or the Internet. ?s 8ell receives an order% it shares production

reuirement information electronically with its suppliers world-wide for

immediate delivery to a 8ell production facility% where the computer is

assem&led and shipped directly to the customer within a week. The 8ell model

relies on demand side pull rather than supply side push - no computer isproduced unless there is corresponding demand in the marketplace. Thus the

massive ueues of inventory usually sitting idle within retail stores% distri&utors%

and factories are virtually eliminated. The productivity advantages of this

production model are profound. 8ell is a&le operate with half the num&er of

employees and one-tenth of inventory of its traditional computer competitors.

Return on invested capital reached 1L; per cent in 1LLL% compared to 1+-0+ per

cent for traditional manufacturing Arms. ,ompanies from around the world have

&een 6ocking to ?ustin% TeGas to understand the 8ell production model% much as

Arms had 6ocked to Tokyo and River Rouge earlier in the century. The opportunity

for productivity improvement was enormous$ in the =S? alone% the cost of goods

in inventory of all value systems was nearly 1 trillion in 1LL. ?s the 1LL+s


7/20

closed% the O8ell modelO &egan to spread from high technology to traditional

manufacturing sectors such as automo&ile production. Recently% Eeneral Motors%

Ford% and 8aimler ,hrysler announced they were moving to electronic supply

chain management systems similar to 8ell ,omputer. If successful% the 8ell

Model could &e every &it as revolutionary to the production structure as FordOs

vertical integration and ToyotaOs lean production models were in earlier eras.

8ellOs originality lay in the approach that it adopted in implementing the direct

&usiness model. In particular% unlike other computer manufacturers% 8ell sells

directly to all of its customers and not Dust to large corporate clients. Through

developing a direct relationship with all of their individual clients and &uilding its

computers to order% 8ell was a&le to &uild a highly eQcient Dust-in-time process%

eliminating most of its inventory in the process. ? further advantage to the 8ell

approach is the instant% current and continuous market research that it produces.

In knowing eGactly what individual customers want in a personal computer or

computer network% 8ell is a&le to anticipate market demand and shape thetechnological and competitive parameters of the computer industry. 8ell argues

that the direct model creates the most compressed ", supply chain &y

eliminating all intermediaries. Moreover% compared to a traditional supply chain

structure% the direct model can reduce inventory investment &y ;+ to + per

cent.

The concept &ehind 8ellOs drive to reduce inventory ineQciencies Phas nothing to

do with stockpiling and everything to do with informationP. 8ue to its made-to-

order approach% 8ell is a&le to see on a daily &asis if% for instance% customer

preference is shifting to larger ", monitors. The company can also discernwhether this is happening for certain customer segments or across the market.

8ell immediately relays its assessment of this information to its suppliers%

allowing them to adDust their inventory accordingly and rapidly meet demand. It

stands to reason that the more information a company has a&out what a

customer wants and how much heJshe reuires% the fewer inventories the

company needs to maintain.

Fewer inventories mean less inventory depreciation. In an industry such as

computer manufacturing% component prices are constantly falling-- typically 1;-

0; per cent per annum. SiG days of inventory '8ellOs norm( compared with 92

days 'standard at ,ompa( can therefore result in signiAcant cost savings oninputs. Furthermore% reduced stockpiles can o7set the risk of &eing caught with

large amounts of o&solete inventory if technology shifts and there is a transition

to a neGt-generation product - as often happens in high technology sectors. It

therefore comes as no surprise to learn that 8ellOs competitors such as IM and

,ompa are constantly striving to cut their inventory levels &ut have yet to

match 8ellOs success in this area.


8/20

The 8ell 8irect Model and irtual alue ,hain Integration

The notion of OlinkagesO &etween supply chain participants is not new and wastraditionally referred to as Overtical integrationO. =nlike the 8ell model% though%

vertical integration implies ownership of &oth upstream suppliers and

downstream distri&utors. Firms such as Ford ha&itually controlled all elements of

the value seuence% vertically integrating the information% decision% Anancial and

operational dimensions of the strategic supply chain.

The spread of Internet-&ased commerce during the 1LL+s resulted in the

emergence of PvirtualP supplyJvalue chain linkages. This approach was perceived

&y many companies as a way of reali#ing the &eneAts of supply chain integration

while avoiding the perceived negative impact of integrating vertically. yseamlessly integrating supply chain suppliers% manufacturers% distri&utors% and

retailers into a single virtual enterprise serving the customer% companies are

achieving huge competitive advantages. The emergence of the virtual value

chain - or PeGtended enterpriseP - has &rought a&out a rethinking of traditional

supply chain relationships and has fundamentally transformed the nature of

competition.

In addition to the previously discussed inventory and disintermediation costs

saving% there are other advantages associated with the direct &usiness model. In

particular% as Michael 8ell states% you actually get to have a relationship with

the customerO. ? direct link to the individual customer provides a manufacturer

such as 8ell with a wealth of marketing and product development information.

This information ena&les the company to &uild a position of strength relative to

&oth its customers and its suppliers. 4hen that information is com&ined with the

technology of the Internet% it allows a company to develop a revolutionary new

value chain infrastructure and &usiness model.

This is what 8ell has done through its Pvirtual integration of the value chainP

approach. Pirtual integrationP means a &lurring of the conventional value chain

&oundaries and roles &etween suppliers% manufacturers and end users. Michael

8ell deAnes Pvirtual integrationP as Pthe idea of interweaving distinct &usinesses

so that our partners are treated as if theyOre inside our companyP. This results in

gains of eQciency and productivity% as well as signiAcant gains in return to

investors. Cigher returns on investment are gained &y concentrating resources

on activities where value can &e added for the customer and not in activities that

simply need to &e done. y this logic% 8ell argues that a computer company% for

instance% does not have to actually make computers. If fa&ricating

semiconductor chips or even placing them on mother&oards does not result in

signiAcant proAt margins% then the computer company should consider

outsourcing such activities. In 8ellOs case% this meant focusing instead on its

distinct core competency - delivering solutions and systems to customers.

8ellOs virtual integration is referred as the model of Pthe digital value chainPwhich means the use of technologyJthe Internet to create a faster% more eQcient


9/20

and more 6eGi&le version of the traditional supply chain. 4ithin a digital value

chain% one company serves as the PanchorP% i.e. Pthe power player around which

the digital value chain is organi#ed and often optimi#edP. The Ppower playerP is

identiAed as such &ecause it either provides the maDor share of the value

delivered to the customer$ it is the dominant supplier$ or it is the owner of a

product or service that cannot &e replicated &y any other mem&er of the valuechain. 8ell is a classic digital value chain anchor. Through its control of the

consumer relationship% it esta&lishes the rules and shapes the competitive

dynamics of the value chain.

In 0++1% 8ell ,omputer was the ", market leader in the =nited States% with a

nearly 1* per cent share% a&out three per cent ahead of second-place ,ompa.

Eateway was third with L per cent% followed &y Cewlett-"ackard with * per cent

and IM with per cent. 8ell overtook ,ompa as the =.S. sales leader in the

third uarter of 1LLL after moving past IM into second place during 1LL*.

Remarka&ly% 8ell has &een a&le to replicate its direct sales approach in do#ens of

other countries. 8uring 0++1% 8ell ,omputer was almost eual with ,ompa interms of glo&al ", sales 'each had a&out 10 percent of the market(. IM ranks

third worldwide% with an * percent share that has steadily declined over the last

few years. Since 1LL;% 8ell had &een gaining market share uickly in all of the

worldOs markets% growing a&out three times the 1U< average annual rate of

glo&al ", sales.

8ellOs sales at its we&site 'www.dell.com( surpassed ;+ million a day in early

0++0% up from ; million daily in early 1LL* and 0; million daily in early 0+++.

In its Ascal year ending Nanuary 91% 0++0% 8ell ,omputer posted revenues of over

91 &illion% up from 9 &illion in 1LL2 -- a compound average growth rate of a&out ;+ percent. @ver the same time period% proAts were up from 1;+ million

to 0 &illion-a U; percent compound average growth rate. Since its initial pu&lic

o7ering of common stock in Nune 1L** at *.;+ per share% the company has seen

its stock price split seven times and increase 2+%+++ percentV 8ell ,omputer was

one of the top ten &est-performing stocks on the >S) and the >?S8?W during

the 1LL+s. In recent years% 8ellOs annual return on invested capital had eGceeded

1;+ percent. Cowever% it has &een hit hard

&y an overall slowdown in ", sales that &egin in the second half of 0+++. The

result has &een a sharp fall in 8ellOs stock price and a reminder that 8ell is not

immune to the cyclical demand and &rutal price competition that hascharacteri#ed the ", industry since the 1L*+s.

The ta&le &elow shows the competitive advantage of supply chain

management of the 8irect model &y 8ell. The selling and administration and

research and development are low compared to the industry. This is &ecause the

virtual integration that 8ell chose to sell directly to customer and the fact that

8ell considers the strategic alliance as within the &oundary of the company and

thus 8ell herself does not heavily invest in RH8 such as making its own chip

&ecause it is making the money &ut instead adopt the partners who is &est in the

Aeld. Inventory turnover is eGtremely high% U2 per year% compared to other


10/20

companies. This is the conArmation of the advantage gaining from the 8ellOs

8irect Model.

If )R" software does not At the model% we donOt want it:

The sentence a&ove &est descri&es what happened at 8ell ,omputer in the early

implementation of )R". From the a&ove discussion% the Pvirtual integrationP of

8ell is the core competitive advantage and it will &e accomplished with or

without the eGistence of )R" software. 8ell ,omputer aggressively gained

market share and already succeeded even &efore the implementation of )R"

system &y selling via telephone or using we& for the Arst to sell computer via

internet. )R" Dust makes it more smoothly to facilitate the implementation of the

8irect model: the philosophy of the company. Therefore% there is no reason that

8ell will consider changing its uniue &usiness model Dust to At the )R"

implementation and it is against the common wisdom we see in practice.

In 1LL% 8ell ,omputer a&andoned their )R" program only after several months

of detail planning and implementation when they reali#ed that is was

inappropriate in their environment.

8ell ,omputer cancelled their )R" contract in Nanuary 1LL

after spending 11; million dollars 'The original cost of the proDect was

estimated at a&out 1;+ million(. 8ell determined that the system could not deal

with the needed sales volume. ?nalysis had focused on ineQciencies caused &y

multiple home-&uilt% unconnected% information systems that inhi&ited information

6ow across the company. This analysis led them in to choose an integrated suiteof applications. )ven as the decision was &eing made% two 8ell eGecutives were

providing suQcient information to invalidate the )R" decision.

?t one of their "latinum ,ouncil meetings where 8ell eGecutives meet with key

customer account ,I@Os% !evin Rollins% 8ellOs ice ,hairman% talked a&out the

critical need for every aspect of the company to &e capa&le of changing its

process rapidly. Ce referred to this as an essential part of what he called velocity

or the continuous speeding up of every &usiness process. ?t that same meeting%

Michael 8ell descri&ed his &usiness as &eing a virtually integrated system of

processes and products eGtending from suppliers through 8ellOs manufacturingand distri&ution processes% on to end customers and the support of the product

on their desktops. Ce also talked a&out the companyOs distri&uted management

style and how continuous process improvement was a way of life throughout the

company.

4hat makes 8ellOs Supply ,hain runB:

,onsider: The company manufactures more than ;+%+++ computers every day%

&ut carries only four daysO worth of inventory% when many of its competitors

carry &etween 0+-9+ days of inventory. Roughly half of its annual revenues

approGimately 1U &illion come from online sales.4hatOs more% 8ell has &een uite forthcoming as to identifying the Psecret sauceP


11/20

responsi&le for its direct model success% itOs the supply chain. 8ell has a single-

minded dedication to supply chain eGcellence% which means esta&lishing

relationships so tight with its suppliers and vendors that they &order on

incestuous in the words of chairman and ,)@ Michael 8ell% P!eep your friends

close% and your suppliers closer.P The company keeps what it calls a supplier

report card on every supplier% and tracks each supplierOs performance against aset of metrics maintained &y 8ell.

?gainst the charge that 8ellOs remarka&ly low inventory levels come at the

eGpense of its suppliers% 8ick Cunter% vice president for 8ellOs ?mericas

Manufacturing @perations% points out% P?&out 9+ suppliers provide ;< of our

direct material purchase spend and most of them maintain eight to1+ days of

inventory in near&y% multi-vendor hu&s. If those levels eGceed 1+ days% we work

with suppliers to lower them since eGcess and o&solete components are not

accepta&le to 8ell% our suppliers or customers.P 8ell also works with its supplier

to prevent inventory levels from &ecoming too low% Cunter adds. PFor 8ell andour suppliers% information is increasingly replacing inventory% and we are

regularly identifying% gathering and sharing new types and levels of data.P

Trading inventory for information is a key to 8ellOs supply chain success% and in

this day of point solutions aimed at tackling small pro&lems uickly. 8ell again is

proof that following its own course rather than Doining the rest of the pack is the

way to go.

8ell runs what is said to &e the worldOs largest implementation of i0 Technologies

Inc.Os software% running its 8ell-speciAc 8Si0 solution on 10+ servers% managing

more than 0;+ suppliers responsi&le for delivering over 9%;++ components.

8ell took an industry that used to &e make-to-stock and shifted it into make-to-

order. In this industry companies lose a price premium every day. ?nytime they

make to stock% theyOre tying up capital% which is the classic supply chain crime.

They tie up in inventory% &ut even worse they lose price advantage for every day

they maintain on-hand inventory.

?&out half of 8ellOs more than ;+%+++ orders each day come through the

Internet. Those orders 6ow through the companyOs legacy order management

system% which records all the orders and releases them to manufacturing. P4e

schedule production lines in every factory glo&ally every two hours%P CuntereGplains. P4e have no inventory and no warehouses in any of our factories.

Instead% weOre a&le to pull material into our factories &ased on actual orders. P4e

literally push a &utton and two things happen%P Cunter continues. P4e lock in the

schedule &y actual order and order num&er into the factory. ?t the same time we

send a message over the Internet to our third-party logistics providers% supplier

logistics. These hu&s have L+ minutes to pull material out of the racks% and

deliver it to 8ellOs &ack door.

To make that happen% 8ellOs ?ustin% TeGas- and >ashville &ased hu&s use

technology from 9 Systems for in&ound% out&ound and inventory management.

This 9 order allocation application supports 8ellOs individuali#ed consumptionproAles% as orders come in from anywhere phone% we&% e-mail% etc. The software


12/20

talks to the 8Si0 system thanks to enterprise application integration software

from we&Methods Inc.% and the systems work together to optimi#e the assem&ly

lines. The solution helped reduce inventory 6ip from 0+ days to siG.

8ell also employs i0 modules to communicate materials reuirements and to

schedule the factories once they receive the components. 8ellOs 4orldwide"rocurement division is responsi&le for negotiating contracts and pricing deals

for all material consumed &y 8ell glo&ally.

The company teamed with ?ri&a Inc. to overhaul a procurement system that

once reuired completion of a three-part paper form% which involved hand

coding information a&out suppliers% part num&ers and item costs% not to mention

manually collecting as many as 1+ approval signatures. These procedures which

could take weeks to complete were costing the company 11+ per reuisition.

8ell implemented ?ri&a uyer over a seven month period% interfacing the

procurement solution with nearly 0+ of 8ellOs legacy systems% including links to

@racle Financials for purchase order% cost centre and accounting code data. Theresult called 8ell Internet Reuisition Tool provides automated processing of fully

validated orders. The system reduced the time to complete a reuisition &y U0<

and the cost &y U1


13/20

drive at the same price as a U+-gig hard drive if the smaller drive &ecomes a

supply issue. 8ell can o7er a promotion on the Internet literally within a couple

hours of reali#ing it has a demandJsupply pro&lem. 8ell has &ecome an eGpert at

ensuring it has the right components at the right time% and it knows where to go

when the usual sources dry up. "artMiner% for eGample% is a supplier of hard-to-

And electronic components to 8ell and its contract manufacturers. The companyhelps meet unplanned upside demand for components essential for production.

In addition% 8ell utili#es "art MinerOs ,?"S data&ase information to manage risk in

the supply chain within its component engineering and procurement groups.

8ellOs reputation for eGcellent customer service eGtends to its reverse logistics

operation as well. The companyOs ?mericas Service 8elivery division% for

instance% uses 4orld ,hain Inc.Os >etwork Repair ogistics solution to handle

roughly 1 million a day in repair volume. The solution provides total visi&ility of

inventory throughout the repair cycle% including real-time and historical reporting

on supplier and repair vendor performance. 4ithin 9+ days of the rollout theaverage repair cycle time dropped from 29 days to 02 days% and today the time

is down to a&out 1 days.

@n the Aeld services side% 8ell uses a supply chain event management solution

from iewlocity Inc. to monitor and manage the return of non-functioning

products% coordinated with a third-party logistics provider. =ltimately% all of 8ellOs

supply chain activities aim at improving visi&ility. The more 8ell knows a&out the

capa&ilities of the supply chain and our suppliers% the &etter decisions 8ell can

make for our customers.

i0X TradeMatriGTM Elo&al Implemation:

8ellOs highly eQcient &uild-to-order &usiness model ena&les customi#ed

computer systems to &e &uilt and delivered uickly. ?s part of the companyOs

continual e7orts to improve its supply chain processes% 8ell recently deployed

i0X TradeMatriGTM supply chain tools to provide glo&al views of forecasted

product demand and materials reuirements% as well as improved factory

scheduling and inventory management. 4ith the new supply chain tools% 8ell

suppliers and Supply ogistic ,enters 'S,s(% or hu&s% have a glo&al view of

short- and long-term materials reuirements in each 8ell factory. 8ell employees

can view materials reuirements% forecasts% and factory schedules and generate

key reports from real-time and near real-time data. The glo&al i0 deployment included software tools% servers% and storage to

support glo&al supply planning and demand fulAlment functions. "latform

selection was a key component in the deployment. ?fter a comprehensive

&enchmarking and selection process% 8ell chose to run its entire i0 supply chain

system on a 8ell and 4indows >TX platform.

The solution leverages the strengths of the 4indows >T operating system% the

MicrosoftX ,luster Server 'MS,S( tool% the Microsoft 4indows oad alancing

Service '4S( tool% and the @racleX Fail Safe '@FS( facility% coupled with the

high performance and scala&ility of 8ell servers% to create a relia&le and cost-

e7ective supply chain infrastructure.


14/20

The i0 solution connects 8ell partners

8ell selected i0 TradeMatriG supply chain management tools to improve

inventory planning% forecasting% and eGecution systems. @n the glo&al supply

planning side of its operation% the company runs TradeMatriG Supply ,hain

"lanner 'S,"(% which creates supplier materials forecasts. The i0 ?ctive 8ata

4arehouse and Metadata 4arehouse support the S," module and 8ellOsreporting solution. rioX and ?ctive Server "ages '?S"( custom reports

supplement the reporting solution. TradeMatriG ,olla&oration "lanner 'T,"(

provides a worldwide view of materials reuirements.

@n the demand fulAllment side% 8ell runs TradeMatriG Factory "lanner 'F"(% a

single-threaded% memory-resident tool% to develop &uild schedules for factories

&ased on current demand priorities% factory capacities% and materials positions. ?

second instance of T," communicates the factoriesO immediate materials needs

to S,s.

These i0 tools were deployed for each operational region% with some components

residing in ?ustin and others in regions around the world. ?ll of the worldwideview reporting tools reside in ?ustin. ?pproGimately 0;+ of 8ellOs largest

suppliers will &e using these tools. Most of these suppliers conduct &usiness with

the S,s% which provide inventory management and are located near the 8ell

factories to ensure timely delivery. These S,s then orchestrate materials

reuirements with 8ell. The i0 tool set connects all three partners in this

relationship.

Selecting the platform

The platform selected to run the i0 modules needed to support current and

predicted 8ell volumes% meet uptime needs% &e highly availa&le and recovera&le%

support 6eGi&le maintenance windows% and scale suQciently. For the initialdeployment in the ?mericas region% the deployment team evaluated platforms

&ased on 4indows >T on 8ell hardware and =>IYX on SunX hardware. The

four general platform types included:

B ? 8ell and 4indows >T platform

B ? =>IY-&ased platform running most data&ase and application components

on =>IY

B ? platform running data&ase components on =>IY com&ined with 4indows

>T upper tiers

B ? platform running the i0 S,"% F"% and data&ase components on =>IY with

the other 4e& and transactional components on 8ell and 4indows >T The teamOs goal was to eGamine the performance and capacity of each i0 module

in each platform environment and recommend hardware platforms &ased on

empirical test results and assumptions a&out future production needs. @nce each

module was eGamined% the complete architecture could &e designed and tested.

In this Arst phase of &enchmarking% the team evaluated each hardware platform

&ased on the following criteria:

B The performance of the i0 modules on various conAgurations of 8ell and Sun

euipment

B "otential compati&ility issues &etween i0 modules and software products

including 4indows >T% @racle% NRun% Nava 8ata&ase ,onnectivity 'N8,( !ona for@racle% and other tools used &y 8ell


15/20

B ?vaila&ility% recovery% performance% cost-e7ectiveness% time-to-delivery%

resource needs% and scala&ility

B Stress points in the application architecture

B "reparation for future regional and worldwide rollouts

)valuating module-&y-module

The team decided early in the process to use 4indows >T and 4S for the 4e&-&ased functions in order to align with 8ellOs eGisting internal system% which

provides Arewall security and Internet portals. The team focused on evaluating

4indows >T and =>IY systems for each of the non-Internet-&ased i0 modules

and non-4e& tiers.

i0 tools and application &usiness data models simulated client activity at the 4e&

and application layers% while other data&ase tools simulated data&ase IJ@ load.

These models were not the Anal ones deployed% &ut provided the closest

simulation availa&le at the time.

First% the team tested components including the data&ase layers and the

memory-resident engines on the Sun )nterpriseTM ;+++ server and the 8ell"ower)dgeX U9++% U9;+% and *2;+ servers. They also tested the application

and 4e& layers &y adding and removing servers to evaluate the performance

and scala&ility% and to locate &reaking points in the architecture tested.

Factory "lanner and Supply ,hain "lanner are &oth single-threaded% memory-

resident applications. For these modules% the di7erence in performance &etween

the =>IY and 4indows >T systems was unsu&stantial in our test cases. To test

Factory "lanner on 4indows >T% the team &egan &y running twice the volume of

8ellOs largest factory. The 4indows >T platform handled three plants at this

volume. ? maGimum of three concurrent F" engines could run on a "ower)dge

four-way server without degradation. The team concluded that Factory "lanner runs e7ectively on 4indows >T% &oth at

the engine and data&ase levels% on separate servers. The F" engine and the

data&ase could run on one server% &ut 8ell internal support team standards do

not allow an application and data&ase to run on a single server. Moreover% the

team felt conAdent that multiple F" data&ases can run on one data&ase cluster

under MS,S and @FS% providing a high level of failover protection.

For the TradeMatriG ,olla&oration "lanner instance for the S, solution% the team

tested 8ell four-way and eight-way servers and a Sun =>IY server for the

data&ase layer. The application and 4e& layers were si#ed for 4indows >T. The

team generated production-euivalent test &eds plus a factor of 1;+ percentmore for growth on &oth the transactional and &atch uploadJdownload

capa&ilities of the T," tool.

@n the glo&al supply planning side% the team performed side-&y-side tests of

Supply ,hain "lanner on 4indows >T and =>IY systems and found that running

4indows >T was noticea&ly faster than =>IY. The team also tested a T,"

instance for the suppliers and found similar results to the T," S, testing% eGcept

that the hardware needed to &e si#ed larger. 4indows >T is strong in memory

management% so the S," i0 code on 4indows >T may &e taking advantage of

this &etter than on =>IY% eGplaining part of our test results. Cowever% another

model or di7erent volumes 'other than the 8ell test models( might have showndi7erent results. This is the nature of the S," and F" tools% which are designed to


16/20

&e conAgured or modeled.

?fter completing the &enchmark testing% the team chose to run all the i0

modules on 8ell hardware running 4indows >T. This architecture provides high

performance and scala&ility. The 8ell and 4indows >T architecture provides the

opportunity for maintenance and upgrades without system downtime. Since

most of the i0 modules and instances can &e uncoupled% components can &eadDusted if necessary.

8eploying i0 glo&ally

8eployment of the i0 solution &egan with demand fulAllment and glo&al supply

planning in the ?mericas 'eGcluding ra#il( in the Arst half of 0+++% then moved

on to regional deployments of demand fulAllment and glo&al supply planning in

?sia and ,hina and glo&al supply planning in )urope in the second half of the

year. In 0++1% 8ell is focusing on deployment of demand fulAllment for )urope

and enhanced functionality for all the modules worldwide.

)ach region was reuired to complete key prereuisite proDects &efore the new

system could &e deployed% including esta&lishing an operational data store tocentrali#e legacy information and a sales forecasting system to work with the i0

solution.

For demand fulAllment% the team decided to deploy Factory "lanner and the

instance of TradeMatriG ,olla&oration "lanner used to communicate with the

S,s in each region. @ne advantage of a modular architecture is that additional

instances can &e deployed if necessitated &y regional growth.

Cardware infrastructure

For the TradeMatriG ,olla&oration "lanner component for demand fulAllment% the

infrastructure includes two "ower)dge U9;+s running 4S to support the

application layer% two "ower)dge U9;+s using MS,S and @FS for the data&ase%"oweraultX U;+F and U9+F storage% and an additional "ower)dge U9;+ as a

&atch uploadJdownload server. The 4e& layer of four "ower)dge 02;+ servers

running 4S supports T," for &oth demand fulAllment and glo&al supply

planning.

The infrastructure for Factory "lanner includes "ower)dge U9;+s using a hot-

swap &ackup method for the F" engine layer and two "ower)dge U9;+s and a

"owerault U;+F using MS,S and @FS for the staging data&ases.

@n the glo&al supply planning side% TradeMatriG ,olla&oration "lanner is powered

&y four "ower)dge U9;+s using 4S for the application layer% two "ower)dge

*2;+s using MS,S and @FS for the data&ase% a "owerault U;+F and U9+F forstorage% and a "ower)dge U9;+ as a &atch uploadJdownload server.

The Supply ,hain "lanner function uses a "ower)dge U9;+ as the engine. The

?ctive 8ata 4arehouse data&ase includes a com&ination of "ower)dge U9;+s

and "owerault U;+Fs and U9+Fs for storage under MS,S and @FS. The Metadata

4arehouse uses two "ower)dge *2;+s using MS,S and @FS% &acked &y a

"owerault U;+F and Ave "owerault U9+Fs for storage. Figure &elow illustrates

this architecture as it was deployed in the ?mericas region.

i0 Factory "lanner and the integration with the eGisting systems:

?s part of a larger deployment of i0 TradeMatriG solutions% 8ell implementedFactory "lanner% a supply chain management module in the TradeMatriG suite.


17/20

Striving to enhance and &roaden the 8ellX 8irect &usiness model% 8ell recently

deployed a suite of software from i0 Technologies. The proDectOs goals were to

&oost the eQciency of 8ellOs supply chain and production planning and eGtend

the scope of 4e&-ena&led information availa&le to factories and suppliers.

Typical of such proDects% 8ell needed to overcome challenges such as constrained

timelines% limited resources% and investments in legacy systems with esta&lishedinterface reuirements.

Factory production work6ow

To provide relevant information and optimi#ed production schedules% the i0

solutions needed to adapt to the eGisting production work6ow% procedures% and

IT systems in 8ell factories. This production work6ow &egins as orders are

received and assigned to factories and eGtends to the picking of parts 'kitting(%

&uilding% and &oGing of ordered systems on the factory lines.

8ell assigns orders to factories &ased on two criteria: geography and product

type. Eeographic assignments primarily aim to minimi#e transportation

eGpenses% &ut also involve considerations such as importJeGport laws andcountry-speciAc product reuirements. The second criterion% product type%

results from factories that speciali#e in producing a single type of product:

desktops% laptops% servers% or storage su&systems.

@nce orders have &een assigned to a factory% the challenge &ecomes weighing

what should &e &uiltBor orders on handBagainst the availa&le supply of parts and

manufacturing capacity.

The availa&ility and delivery of materials can complicate the decision of eGactly

what to &uild. Independently owned and operated Supply ogistics ,enters

'S,s( or hu&s usually deliver materials to factories. ?n S, coordinates the

delivery of components to maintain a timely% damage-free% and controlled 6ow of supplies to the assem&ly line. Figure &elow illustrates the factory production

work6ow% from the delivery of materials through the shipping of Anished

products.

Figure . Factory production work6ow at 8ell

S,s help improve eQciency and reduce inventory% &ut also introduce planning

challenges. ?n S, can service multiple factories$ each factory makes materials

reuests independently of the others. For planning purposes% each factory

assumes an unlimited supply at the S, in making materials reuests. This

procedure allows each S, to track materials availa&ility% determine whichreuests it can and cannot meet% and commit 'or refuse( to deliver the reuested

materials.

8ellOs eGisting IT infrastructure

The IT landscape at the &eginning of the proDect included Ave maDor applications

that needed to integrate with Factory "lanner. Figure &elow illustrates the

architecture of these Ave systems.

Figure. )Gisting manufacturing IT systems

8ell @rder Management System '8@MS(. This legacy application records all

orders and releases them to manufacturing. ?ccessed via character-&asedterminal screens% several distinct systems actually fulAll 8@MS functions%


18/20

depending on the geographic region and type of order. 8@MS manages several

di7erent types of transactions% from individual credit card transactions to

&usiness purchase orders. The largest customers freuently create &lanket

purchase orders and use "remier 8ell.com to select products from a list of

approved conAgurations. These orders are internally approved and then sent to

8ell via the Internet.4ork-in-"rogress Tracking ,oding System '4T,S(. This system lies at the heart of

the manufacturing process. Moving an order from 8@MS to 4T,S releases it to

manufacturing. 4T,S creates a &ill of materials '@M( and a uniue service tag

num&er for each system. The @M and service tag num&er are printed on a

production order called a Ptraveler%P which follows the order through the

production process. The service tag num&er is also &urned into each product and

used for warranty and maintenance purposes.

)ngineering Materials "rocess and ,ost Tracking ')M"?,T(. This inventory control

program governs on-hand inventory. Eenerally% )M"?,T tracks small parts that

are ordered in uantity rather than delivered on demand &y an S,. These partsrange from ineGpensive screws to IntelX processors.

alue ,hain 'valuechain.dell.com(. This eGtranet portal for suppliers provides

direct% Internet-&ased access to 8ell documentation% forms% tools% and supplier

rating scorecards.

@perational 8ata Store '@8S(. This data&ase is optimi#ed to support nearly real-

time manufacturing decision support ueries. Typical ueries might ask the

location of a particular order on the manufacturing 6oor or what systems

compose the order of a particular customer. Materials-related ueries could

include the uantity of a part on hand or in transit from an S,. ecause

production plans were previously &ased on this data% the data used as input &yFactory "lanner is already Pcleaned.P

Several characteristics made @8S a successful tool. First% it contains only

information a&out the current dayOs production. ?lthough su&stantial e7ort is

made to update data in near real time% in practice% periodic updating is suQcient

for successful operation. Second% the granularity of the @8S data is the same as

the data in the transaction systems. @ther true data marts aggregate data and

integrate it with historical data for forecasting% sales trend analysis% and other

ueries across time. >eGt% the user interface supports several means of

accessing data% such as reports and spreadsheets. Finally% &ecause @8S is

optimi#ed as a read-only data&ase% only a minimal num&er of operations writenew data.

>ew i0 applications deployed

Factory "lannerJ?dvanced Scheduler% a client-server tool% optimi#es the planning

and scheduling of manufacturing production. 8ell deployed Factory "lanner

version 2. =sers access numerous F" functions through a graphical user interface

'E=I(. ,ompleG &ut eQcient algorithms simultaneously consider materials and

capacity constraints and propose appropriate production schedules. "roviding

proper values to hundreds of PswitchesP tailors the algorithms to an individual

customer% which simpliAes implementation and future upgrades. The resulting

conAguration is referred to as the model. TradeMatriG ,olla&oration "lanner was initially deployed as the work6ow and


19/20

communication application to provide supply chain visi&ility across multiple

enterprises. T," allows manufacturers and suppliers to colla&orate dynamically

on materials delivery agreements. ? three-tiered application% T," consists of a

4e&-&ased E=I% a &usiness logic tier% and a data&ase.

,reating an integrated architecture

To design the architecture% 8ell Arst needed to determine the num&er of Factory"lanner instances to run% or conversely% the num&er and type of orders to &e

handled &y each instance. 8ell decided to run an instance of Factory "lanner in

each factory% in part &ecause orders are routed to factories &ased on general

&usiness decisions% such as product type or geographic location.

The high-performance% cost-e7ective systems now availa&le made implementing

Factory "lanner in each factory possi&le. 8uring the last 0+ years% computing has

evolved from room-si#ed% water-cooled mainframes with 90 M of memory

running at 2+ MC# to multiple servers per rack% each with giga&ytes of memory

running at gigahert# speeds. This level of technology allows managers to use F"

as a decision support tool-they can uickly perform several Pwhat ifP scenarios tosolve particular pro&lems &efore Anali#ing the production plan.

>eGt% the input to Factory "lanner needed to &e assem&led in the reuired 6at

Ale format. 8ata is grouped into three categories:

B consists of orders released to manufacturing% which are sorted &y associated

priorities or the promised or target ship date. 4T,S pulls these orders from

8@MS.

B consists of work in progress from 4T,S% on-hand parts from )M"?,T% and S,

inventory for &oth on-hand and in-transit materials from T,".

B includes the @Ms from 4T,S% the F" plan or model% and other current

information a&out the factory. This factory information covers routings%resources% factories% operations% stockrooms% and &ase type codes used to route

di7erent products on 4T,S.

,reating and implementing an F" model is a signiAcant proDect% involving more

than ;++ 6ags. )Gtensive skills and eGperience are reuired to create an

accurate model of the products% materials% logistics% and manufacturing facilities.

)7ective and accurate modeling% which often reuires the assistance of

eGperienced consultants% can shorten the development time and allow the

modeling to &e done in conDunction with other integration and deployment tasks.

The following steps provide an overview of the Factory "lanner cycle:

1. Snapshots are taken of demand data% structure data% and inventory% &asedon data from the S,s and )M"?,T% which is updated periodically.

0. F" generates an inAnite capacity plan% which Alters out orders with missing

parts and creates an eGception report. =sers can make adDustments &y selecting

di7erent systems to &uild or su&stituting materials.

9. F" generates a detailed schedule with an eGception report that identiAes

capacity and resource-loading issues. =sers can rerun F" repeatedly to resolve

these issues-until the production schedule is satisfactory. ?ny orders left in the

eGception report remain in the system &ut are pushed into future production

cycles.

2. ?fter a particular plan has &een saved in F"% a detailed schedule is sent to4T,S to use on the factory 6oor.


20/20

;. Materials reuirements% or &uild plans% were forwarded to the S,s via T,".

)ach S, picks these up &y accessing an account on alue ,hain% which provides

a single point of record for 8ell and its suppliers. )ach S, then responds with a

commitment to deliver all or a portion of the reuested materials. If an S, has

not responded in the reuired time% a teGt page notiAes production control.

The 8ell deployment of Factory "lanner identiAed several &est practices that helpensure a successful proDect:

B =se Factory "lanner as a decision support system to empower those running

the factory% rather than using it as a transaction system. This is facilitated &y

technology that reuires minutes for actions that once took hours.

B =se data that is Pgood enough.P 4hile sophisticated tools need good data%

data will never &e perfect in a compleG and fast-moving operation. For 8ell% this

was adeuate to produce successful production.

B ,reate a loosely coupled system &y keeping information 6ow as

asynchronous as possi&le. Such a system will &e ro&ust and 6eGi&le as new

components and applications are added.

Trends in Food Innovation

Documents

Transcript of Trends in Food Innovation