International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 7

ISSN 2229-5518

Optimizing the Formula of Composite Non-Rice Carbohydrate Sources for Simulated Rice Grain Production

Iyus Hendrawan, Sutrisno, Purwiyatno Hariyadi, Y.Aris Purwanto, Rokhani Hasbullah

Abstract—Food diversification program is becoming more importance to reduce the dependence on rice as a staple food in Indonesia. Simulated Rice Grain (SRG) made of non-rice carbohydrate sources was expected to be a subtitute of rice. SRG was formulated based on nutritional value and physichochemical properties of local Ciherang rice flour as a standard. Goal Linier Programming (GLP) was used as optimization method to formulate SRG using various non-rice carbohydrate sources, including arrowroot starch, canna starch, sago starch, sugar palm starch, beneng taro flour, white sweet potato flour, tapioca flour, white corn flour, sorghum flour and breadfruit flour. Optimization parameter used were nutritional value (protein,fat,amylose,amylopectin,carbohydrate,ash),and physical properties (color index,bulk density and angle of repose).The result showed that optimum composite non-rice carbohydrate sources for SRG (SRG flour) consisted of arrowroot starch (30 percent), beneng taro flour (42 percent) and sorghum flour (28 percent). The optimum SRG flour had a predicted nutritional value of 11.78 percent of moisture, 1.97 percent of ash, 1.32 percent of fat, 6.22 of percent protein, 1.28 percent of food fiber, 1.74 percent of crude fiber, 1.46 percent of total sugar, 22.52 percent of amylose, 63.48 percent of amylopectin, and physical

properties of 39.01 degree of angle of repose, 68.59 percent degree of color, and 446.21 kg/m3 of bulk density. Our results showed that

these predicted nutritional and physical properties value of SRG flour is similar to that of analyzed values.

Index Terms - Simulated rice grain, various sources of non-rice carbohydrate, goal linier programming, optimization

————————————————————

1 INTRODUCTION

eeting the needs of staple food can be carried out in three ways i.e. farming intensification, land extensifica- tion and product diversification. Intensification is an
effort to maximize land potency by a variety of activities. This condition will certainly reach the optimum point of produc- tion activity both in terms of seed, fertilization, irrigation and land management. Meanwhile, extensification is an effort to meet needs for food by expanding land area. This effort should consider land condition in order to prevent high envi- ronmental risk and cost. Food product diversification is an effort to meet needs for carbohydrate instead from rice.
Indonesia has great potency in terms of carbohydrate sources such as cassava, arrowroot, canna, breadfruit, sweet potato, corn, taro, gembili, suweg, gadung, huwisawu, kimpul, Java potato and sago. With 52 million Ha of forest which man- aged to produce wood, 1,560 million tonnes per year of food stuffs can be produced [1]. Various carbohydrate sources have similar basic component with rice. Thus, they have potency as alternative material sources for rice substituon.
Recently, there is a tendency of decreasing number of rice consumption per capita and increasing number of imported food stuffs such as wheat and increasing number of potatoes

————————————————

• 1Doctor degree program in Agricultural Engineering Science , Bogor

Agricultural University.E-mail : iyushendrawan@yahoo.com

• 2Department of Mechanical and Biosystem Engineering, Faculty of Agricultural

Engineering and Technology, Bogor Agricultural University.

E-mail :Kensutrisnio@yahoo.com

• 3Department of Food Science and Technology, Faculty of Agricultural

Engineering and Technology, Bogor Agricultural University. E-mail :phariyadi@ipb.ac.id

• 4Department of Mechanical and Biosystem Engineering, Faculty of Agricultural

Engineering and Technology, Bogor Agricultural University.

E-mail :arispurwantoo@ipb.ac.id

• 5Department of Mechanical and Biosystem Engineering, Faculty of Agricultural

[2]. This shows that food diversification has actually been im- plemented by Indonesian people. However, this program should also consider with Indonesian people psychology which indicates that eating is represented by eating rice. Eat- ing rice gives its own pleasure and not boring so there is a tendency to consume rice continously and in large amount although it is not healthy enough [3].
An effort in making grain-like rice had been introduced in several grain name, ingredients and technology. Artificial rice is made from various flour source by adding certain nutrient and flavor which not contained in rice and then produced using roll-type granulator [4]. Simulated Rice Grain (SRG) had been made through fortification technique using Ferrous sulfate heptahydrate(FSH) through extrusion process [5]. SRG had also been made from rice flour, iron compound and 25% of water using single screw-extruder[6] and addition of micronutrients [7]. In making grain-like rice, extrusion technology had been employed together with rice flour and 30% of starch [8]. Composition of 70% of maize flour and 30% of starch using extrusion process produced good result of analogue rice [9].
A study on composing formula of SRG flour made from
various non-carbohydrate sources is importantly needed to
produce rice grain materials or enriched rice materials.
A process of producing SRG needs optimization process in order to produce simulated rice with close characteristic to rice. The purpose of this research was to determine the opti- mum formula of composite non-rice carbohydrate sources as a material for SRG production.

2 MATERIALS AND METHODS

The research was conducted for 9 months started from
March untill November 2013 in Laboratory of Food Analysis -
Engineering and Technology, Bogor Agricultural University. E-mail :rohashb@yahoo.com

IJSER © 2015 http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 8

ISSN 2229-5518

Department of Food Science and Technology, Laboratory of Agricultural Product and Food Processing Engineering – De- partment of Mechanical and Biosystem Engineering, Bogor Agricultural University and in Laboratory of Indonesian Cen- ter of Agricultural Postharvest Research and Development, Ministry of Agriculture of Republic of Indonesia.
This research was carried out under these following stag- es: i) Prepare ten non-rice carbohydrate sources and Ciherang

10

∑ a8i xi ≥ a8 st

i =1

10

∑ a11i xi ≥ a11st

i =1

Objective Function

Minimum

(2) (3)

variety of local rice in the form of flour, ii) Evaluate the nutri-
 10

−

  10

+

  10
− (4)
ent content and physical properties of flour made from non-

z = W 4  ∑ a4i xi − a4st  + W 8  ∑ a8i xi − a8st  + W 11 ∑ a11i xi − a11st 

rice carbohydrate sources and Ciherang of local rice variety, iii) Develop mathematic model that will be processed using
 i=1
  i=1
  i=1 
Goal Linear Programming (GLP), iv) Evaluate the nutrient content and physical properties of flours mixture based on the

Conversion into LP model

optimum formula resulted from GLP.
Materials used in this experiment were obtained from local
farmers consisted of arrowroot starch (Maranta arundinacea
Linn.), canna (Canna edulis Ker.), Beneng taro (Colocasia esculan-

ta( L.) Schott), white sweet potato (Ipomoea batatas Poir), tapio-

10

1 4i i 4st i=1

and

10

2 8i i 8st i=1

10

2 11i i 11st i=1

ca flour (Manihotutilissima Pohl.), white corn (Zea mays L.), sa-

y = y − y ; y ≥ 0; y ≥ 0 y

= y − y ; y

≥ 0; y ≥ 0

go (Metroxylonsagu Rottb.) which obtained from Jakarta, sugar

1 1 1 1 1 2

+ − + −

2 2 2 2

palm starch (Arengapinnata Merr), sorghum (Sorghum bicolor
(L.) Moench) Nambru variety, and breadfruit flour (Artocarpus

communis Forst), rice (Oryza sativa L.) of Ciherang variety. All

= − ;

3 3 3

≥ 0; ≥ 0

3

materials were in the form of flour and sifted using 120 mesh
sieve size. Those materials were then analyzed to determine
the nutrient contents and physical properties of the flours.

Minimum

− + _

Moisture, ash, crude fibre content were analyzed using
Gravimetri method, fat content by Soxhlet method, protein

z = W 4 1

+ 8 y

+

11 11

(5)

content by Kjeldahl method, carbohydrate used method of by difference,dietary food fiber content by enzymatic method and total sugar content by titration method. Starch content and proportion of amylose and amylopectin was determined by

Constraint;

10 +

∑ a4i xxi − ( y

− y ) = a4 st

(6)

spectrophotometer method. Angle of repose was measured using AOAC(1984), whiteness degree by whiteness meter and

1 1

i =1

10 + −

bulk density were determined by weighing the sample at
specified volume of glass cylinder[10]

∑ a8i xxi

i =1

− (

2

− ) =

2

a8 st

(7)

Those carbohydrate source materials were formulated to be used as material for simulated rice grains (SRG) production (SRG flour). The formula of SRG flour was then optimized using Goal Linear Programming (GLP) [11], [12].
Eq.1 to Eg.3 were developed based on desired physico-
chemical properties, and Eq.4 as objective function was devel- oped based on defined penalty weight which then subjected to minimization. Convert the objective function which subjected to minimization into linear programming program (Eq. 5). The constraint functions were Eq. 6 (protein), Eq. 7 (amylose con- tent), Eq. 8 (color index), Eq. 9 (moisture content), Eq. 10 (ash content), Eq.11 (fat content), Eq. 12 (food fiber), Eq.13 (crude fiber), Eq. 14 (total sugar), Eq. 15 (amylopectin), Eq.16 (angle of repose), Eq. 17 (density) and the minimum requirement of starch flour ratio was 30 percent [8] (Eq.18).

10 +

∑ a11i xxi − ( y

i =1

10

∑ a1i xi ≥ a1st

i =1

10

∑ a2i xi ≥ a2 st

i =1

10

∑ a3i xi ≥ a3st

i =1

10

∑ a5i xi ≥ a5 st

i =1

10

−

− ) =

3

(8) (9)

(10)

(11) (12)

10

∑ a4i xi ≥ a4 st

(1)

∑ a6i xi ≥ a6 st

i =1

(13)

i =1

IJSER © 2015 http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 9

ISSN 2229-5518

10

∑ a7i xi ≥ a7 st

i =1

10

∑ a9i xi ≥ a9 st

i =1

(14)

(15)

0, z minimum was only determined by W8 . Referring to eq. 6,
7 and 8, the optimum protein was 8.58 percent, surplus of am-
ylose 25.59 percent and surplus of color index was 57.74 per- cent. Using total composition of x1 = 0.6554, x3 =0.9224 and x9 =0.6068, total protein became 6.22 percent, amylose 22.52 percent and color index 68.59 percent.

10 (16)

If W
≥ W and W ≤ W
were not met, (𝑦+ − 𝑦− ) =

∑ a10i xi ≥ a10 st

4 8 8 11 1 1

+ − + − − −

i =1

10

∑ a12i xi ≥ a12 st

i =1

(17)

−4.71,(𝑦2 − 𝑦2 ) = 23.25(𝑦3 − 𝑦3 ) = 41.35, 𝑦1 = 4.71, 𝑦2 = 0
,𝑦− = 0 was obtained, then z minimum was only determined
by W4 and W8 . Referring to Eq. 6, 7 and 8, the protein was 3.87
percent, amylose was surplus 23.25 percent and color index
was surplus 41.35 percent. Using totalcomposition of x1 =

7 x1 + 7 x2

+ 7 x5

+ 7 x7

+ 7 x8

− 3 x3

− 3 x4

− 3 x6

− 3 x9

− 3 x10

≥ 0 (18)

0.5054, and x9 =1.1772, the total protein became 2.31 percent, amylose was 27.86 percentand color index was 79.35 percent.

Non-negativity constraint

xi ≥ 0, i = 1...10

3 RESULTS AND DISCUSSIONS

3.1 Nutrient Content and Physical Properties of Various

Carbohydrate-based Flours

The analysis result of nutrient content and physical prop- erties of arrow root starch (x1 ), canna starch (x2 ), Benengtaro flour (x3 ), white sweet potato flour (x4 ), tapioca flour (x5 ), white corn flour (x6 ), sago starch (x7 ), sugar palm starch (x8 ), sorghum flour (x9 ), breadfruit flour (x10 ) and Ciherang variety of local rice flour (xst ) that were used as coefficient to formu- late the constraint in GLP is shown in Table 1. The nutrient contens and physical properties values will be used as coefficient to formulate the contraint in GLP.
Eq. 5 could produce SRG flour with have close characteristic of protein and amylose content and whiteness degree to Ciherang rice flour. The desired protein content of SRG was 8.58 percent. This value was difficult to obtain as the protein content of SRG material ranged between 0.69 to 8.38 percent. Therefore, another non-carbohydrate source which have higher protein content should be added. The desired amylose content of SRG was 23.61 percent or lower. This value was in the range of amylose content of the material which ranged between 14.92 to 37.3 percent. The desired whiteness degree of SRG was 92.1 percent. This value was in the range of the whiteness degree of raw materialwhich ranged between
52.1 to 93.6 percent

3.2 Model Execution Process using Linear

Programming

The used of linear programming to solve Eq.5 as objective function and Eq. 6 to 18 as constraint functions produced the optimum value (z) for various penalty weight i.e. W4 (penalty weight for protein), W8 (penalty weight for amylose) and W11
(penalty weight for color index) is shown in Table 2.
With z minimum value of 25.59 percent, it produced optimum composite non-rice carbohydrate sources for SRG with the following composition: 0.66 portion or 30 percent of arrowroot starch, 0.92 portion or 42 percent of Beneng taro flour and 0.61 portion or 28 percent of sorghum flour.

3.3 Nutrient Value and Physical Properties of Optimum

Composite Non-rice Carbohydrate Sources for SRG

Optimum flour for SRG found using GLP (30 percent of ar- rowroot starch, 42 percent of beneng taro flour and 28 percent of Sorghum flour) was analyzed for its nutritional value and physical properties. Predicted nutritional values of flour for SRG obtained from optimization result as compared to that of analyzed values, and flour from Ciherang variety of local rice are shown at Table 3 and Table 4.

3.4 Protein and Amylose Content

The SRG flour optimization with composition of 30 percent of arrowroot starch, 42 percent of beneng taro flour and 28 per- cent of sorghum flour by using Eq.6 and minimum z value produced protein content 6.22 percent. This value was still lower than the desired result i.e. 8.58 percent. This could be due to the protein value of the composite materials (3.40± 3.09) percent with large variance. Some raw materials of the composite flour had higher protein content than rice; however Eq.18 indicated that starch content for lower protein content was set up to 30 percent [8].
The optimum SRG flour has amylose content of 22.52 per-
cent while the standard value was 23.61 percent. The amylose content of SRG flour and Ciherang rice is in medium range [3]. The optimum amylose content of SRG was still in the range of its raw material i.e. (28.01±6.05) percent. The simulation result using penalty weight of amylose content higher than penalty weight of protein and higher or similar to penalty weight of color index produced SRG flour with amylose content 27.86 percent; protein content 2.31 percent and higher color index

+ − i.e. 79.35 percent. The resulted composition was arrowroot

At W4 ≥ W8 and W8 ≤ W11 , it was obtained (𝑦1 − 𝑦1 ) =

+ − + −

− − −

0,(𝑦2 − 𝑦2 ) = 25.59, (𝑦3 − 𝑦3 ) = 57.74, 𝑦1 = 0, 𝑦2 = 0 ,𝑦3 =
starch 30 percent and sorghum flour 70 percent.

IJSER © 2015 http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 10

ISSN 2229-5518

Table 1. Physicochemical properties of various carbohydrate sources

Nutrient con- tent/physical properties of starch/flour

Flour material

x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 xst

Moisture (%,wd) (a1i) Ash

(%,db) (a2i) Fat (%,db)(a3i) Protein (%,db)(a4i) Food fiber (%,db)(a5i) Crude fiber (%,db) (a6i) Total sugar (%,db) (a7i) Amylose (%,db)(a8i) Amilopectin (%,db)(a9i)

Angle of repose

(degree)(a10i) Color index (%)(a11i) Density (kg/m3 )(a12i)

(9.9

± 0.19)

(0.27

± 0.03)

(0.36

± 0.00)

(0.65

± 0.09)

(2.67

± 0.23)

(0.49

± 0.01)

(1.03

± 0.30)

(28.55

± 0.93)

(65.98

± 0.79)

(35.1

± 0.44)

(83.6

± 0.05)

(514

.±10.5)

(16.86

± 0.12)

(0.20

±0.01)

(0.45

±0.15)

(0.69

±0.07)

(2.38

±0.15)

(0.57

±0.04)

(1.47

± 0.07)

(37.3

±0.29)

(56.68

± 0.51)

(45.27

± 3.04)

(72.67

± 0.05)

(498

± 5.26)

(11.93

± 0.14)

(4.32

± 0.06)

(0.9

± 0.03)

(6.86

±0.08)

(2.47

± 0.10)

(3.24

± 0.02)

(2.00

± 0.05)

(14.92

± 0.35)

(65.31

± 0.21)

(34.27

± 0.05)

(52.05

±0.05)

(396.32

± 0.09)

(7.26

± 0.01)

(1.96

± 0.08 )

(0.59

± 0.05)

(5.52

± 0.23)

(2.34

±0.14)

(2.57

± 0.01)

(4.32

± 0.18)

(25.28

± 0.20)

(57.43

± 0.42)

(32.5

± 0.33)

(70.5

± 0.00)

(487.2

± 3.02)

(4.62

± 0.01 )

(0.06

± 0.00)

(0.29

± 0.01)

(0.46

± 0.00)

(1.52

± 0.07)

(0.37

± 0.03)

(1.09

± 0.04)

(29.54

± 0.25)

(66.68

± 0.01)

(25.34

± 4.86 )

(93.6

± 0.05)

(467.7

± 0.47)

(3.60

± 0.10)

(0.49

± 0.04)

(2.03

± 0.07 )

(8.38

± 0.13 )

(3.16

± 0.19 )

(0.32

± 0.09)

(2.21

± 0.10)

(24.11

±0.52)

(59.30

±0.66)

(49.16

± 1.14)

(82.5

± 0.00)

(399.08

± 5.86)

(14.59

± 0.04)

(0.23

± 0.04)

(5.58

± 0.05)

(5.36

± 0.05 )

(1.50

± 0.06

(0.41

± 0.01)

(0.32

± 0.11)

(32.99

±0.36)

(53.60

± 0.36)

(41.47

± 0.65 )

(59.15

± 0.24)

(498.68

± 4.09)

(12.57

± 0.01)

(0.22

± 0.06 )

(0.47

± 0.01 )

(0.66

± 0.00)

(1.74

± 0.15)

(0.48

± 0.06)

(1.33

± 0.13)

(31.99

± 0.58)

(63.11

±0.48)

(40.08

±0.01)

(70.7

± 0.12)

(540.86

± 1.21)

(11.28

± 0.10)

(0.54

±0.02)

(0.96

±0.02)

(6.39

± 0.04)

(4.65

± 0.25)

(0.80

± 0.02)

(1.10

± 0.13)

(27.57

±0.19)

(58.34

± 0.32)

(50.46

±1.00)

(77.53

± 0.10)

(448.5

± 1.72)

(9.03

± 0.13)

(3.47

± 0.41 )

(4.34

± 0.15)

(5.83

± 0.03 )

(2.47

± 0.21)

(0.54

± 0.06)

(1.69

± 0.09 )

(23.28

± 0.46)

(58.38

± 0.86)

(40.16

± 0.54 )

(69.08

± 0.30)

(367.5

± 3.07)

(11.08

± 0.00)

(0.33

± 0.08 )

(0.43

± 0.03)

(8.58

± 0.01)

(6.88

± 0.17)

(0.32

± 0.02 )

(1.16

± 0.16 )

(23.61

± 1.21 )

(58.69

± 0.99 )

(42.85

± 0.99 )

(92.13

± 0.13 )

(467.47

± 2.09 )

IJSER © 2015 http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 11

ISSN 2229-5518

Table 2. Optimum value for various penalty weights

The minimum and maximum value of angle of repose re- sulted from ten materials which were used for optimization were (25.3±4.86) degree and (50.5±1.00) degree with average
number was (39.38±7.80) degree. Optimizing angle of repose was carried out to determine the optimum value of angle of repose of the mixture. This was expected that SRG flour could flow properly when fed into the moulding machine. The op- timum angle of repose was 39.57 degree and testing result was
32.9 degree. The angle of repose of Ciherang rice was 42.85 degree at forming machine. It was higher than the optimiza- tion result or testing result of SRG. This condition made the flow of mixture could have better performance if using the angle of repose of Ciherang rice.
The minimum and maximum values of color index of ten materials were (52.05±0.05) degree and (83.60±0.05) degree
with average number was (73.13±12.03) degree. The color in- dex of flour made by Ciherang rice was (92.10±0.13) percent. Facing this condition, it would be very difficult to meet the
color index standard as the number was out of ranges. The optimum color index of SRG flour was 68.59 degree and la- boratory test result was (59.96±0.04), which were still lower than expected result.
The minimum and maximum density values of ten materi-
als were (367.5±3.07) kg/m3 and (540.85±1.21) kg/m3 with aver- age number was (461.82±57.38) kg/m3. The density of flour made by Ciherang rice was (467.47± 0.209). The optimum density resulted from optimization was 446.21 kg/m3 and la- boratory test result was (455.0±0.00) kg/m3. It was expected that density of SRG flour close to the density value of Ci- herang rice flour.

IJSER © 2015 http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 12

ISSN 2229-5518

Table 3. Predicted nutritional value of SRG flour obtained from optimization result as compared to that of analyzed values,
and flour from Ciherang variety of local rice

Component Predicted

value

Analyzed value Ciherang variety of local rice

Moisture (%,wb) 11.78 (8.65±0.04) (11.08± 0.00) Ash (%,db) 1.97 (0.63±0.02) ( 0.33± 0.08) Fat (%,db) 1.33 (1.42±0.00) (0.43±0.03) Protein (%,db) 6.22 (8.30±0.11) (8.58±0.01) Carbohydrate (%,db) 90.48 (89.65±0.38) (90.66±0.02)

Dietary food fiber (%,db) 1.28 (2.63±0.19) (6.88.±0.12) Crude fiber (%,db) 1.74 (0.55±0.01) (0.32±0.02) Total sugar(%,db) 1.46 (0.76±0.11) ( 1.16±0.16)

Starch ( % ,db) 86.00 (85.70±0.20) (82.30±0.22)

Table 4. Predicted physical properties of SRG flour obtained from optimization result as compared to that of values analyzed, and flour from Ciherang variety of local rice

No Physical properties Predicted value Analyzed value Ciherang variety of local rice

1 Angle of repose (degree) 39.01 (32.89±0.61) (42.85± 0.99)

2 Color index ( %) 68,59 (59.96±0.04) ( 92.13± 0.13)

3 Density ( kg/m3) 446.21 (455±0.00) (467.47±2.09)

4 CONCLUSION

Linear Programming method could be used in optimiza- tion process of simulated rice grain production by considering objective function (protein, amylose and color index) and con- straint function. SRG flour formulated using 30 percent of ar- rowroot starch, 42 percent of Beneng taro flour and 28 percent of sorghum flour has similar physicochemical properties to that flour from Ciherang variety of local rice. Future research is still needed to explore various non-rice carbohydrate sources as an effort to produce closer rice physicochemical properties.

ACKNOWLEDGMENT

The authors would like to say thank to Doctoral Grant from
Ministry of Education and Culture of Republic of Indonesia and
also National Development and Research Partnership program from Center of Agricultural Research (KKP3N) – 2013.

REFERENCES

{1]Suhardi, Sabarnurdi S., Soedjoko, S.A, Dwidjono, H.D., Minarningsih and Widodo, A. (1999). Forest and Estate as National Food Source, Department of Forestry, Jakarta(ID).

[2]Rangkuti, P.A. (2009). Communication Strategy on Food Security

Development). J.Lit.Pert 28(2). 39-45.

[3] Haryadi. (2008). Rice Processing Technology. Gadjah Mada University

Press, Yogyakarta(ID).

[4] Kurachi, H. (1995). Process of Making Enriched Artificial Rice .US. Patent

N0 5,403,606.

[5]Kapanidis, A.N and Lee, T.C,. (1996). Novel method for tht production of color-compatible ferrous sulfate-fortified simulated rice throught extrusion. Journal of Agricultural and Food Chemistry, 44(2), 522-525.

IJSER © 2015 http://www.ijser.org

International Journal of Scientific & Engineering Research, Volume 6, Issue 3, March-2015 13

ISSN 2229-5518

[6]Moretti, D., Lee, T.C., Zimmermann, M.B., Nuessli, J, and Hurrel, R.F,. (2005). Development and evaluation if iron-fortified rice grain. Journal Food Science.70(5).S330-S336.

[7]Bruemmer T, Kleemann N, Meyer M, Schweikert L.(DSM and Buhler A,G). (2005). Rice-based food compositions and processes for their preparation. WO patent application 2005/053433 A1.2005-Jun-16.

[8]Mishra, A., Mishra, H. N., and Rao, P.S. (2012). Prepration of rice ana- logues using extrusion technology. International Journal of Food Sci.

&Tech 1-9.

[9] Budi, F.S., Hariyadi, P., Budijanto, S. and Syah, D. (2013). Extrusion process technology of analog rice. Pangan.3 September 2013;263-274.

[10} Hartoyo,A. and Sunandar,F.H. (2006).Utilization of Composite Flour from White Sweet Potatoes (Ipomoea batatas L), Germinated Soybeans (Glycine max Merr.), and Germinated Mung Beans (Virginia radiata L) as Wheat Flour Partial Substituent of Alternative Food, High Energy-Protein Biscuit. J Teknol.Indust. Pangan XVIII(1) . 50-57.

[11]Prasetyo, A. K.(2002), Guidance on QM Application Program for Window

Version 3.0. ). PT Elex Media Komputindo Jakarta (ID ).

[12]Siringoringo,H. (2005). Operational Research Technique for Linear

Programming.Penerbit Graha Ilmu. Yogyakarta(ID).

IJSER © 2015 http://www.ijser.org