Greenhouse design – TEP 400

November 27th, 2012

 

it’s the design about greenhouse, it’s my homework from my lecturer so i have to learn again and try to do that. it’s for tomato crops.

The problem for this homework is you must design the greenhouse for 1000 tomato crops and you must plant it in polybag (35×40 cm). We must design the greenhouse for that crops  include the size, structure, how we design and take the polybag in the right place and we get the optimization. The distance between crops is 50 cm, and the distance between the rows is 100 cm.

With the information from book and journal, we will get many information for completing this homework. We can use the Lagrange Multipliers to get the optimization for know how many rows and columns for this greenhouse.
I tried to calculate it with Langrange Equation and i got how many rows and columns to optimize for locating the polybag. the rows were 14 and the columns were 72. With the equation, we will get the total length and total width for the greenhouse. Total length was 3790 cm and total width was 1990 cm. From the journal from Suhardiyanto (2007), the great design for Indonesia is modified Standard peak greenhouse.

Rantek 400

Script

PDF

Hopefully it will give u a benefit to look for this or download it.

For Mr. Kudang, thanks for this homework because it give me a reminder to study again for Engineering Design..

note : some of the scripts were missing because if u close the AutoCad 2012 the scripts will be disappeared.

Regards,

Sandro Pangidoan

 

My group project

November 14th, 2012

 

 

it’s my group project for our subject in my major. the name of this design is “Alat Penyayat Ikan” or “The Filler of Fish”.

This design started from the problem to get the fillet from fish but it’s really hard for us if we don’t have an ability like chef. it will waste our time and some of us will think it’s great to buy it in supermarket.

the other problem is nowadays many people prefer to eat fish without the bone because it will be busy and waste our time if you must be boning the fish when you want to eat the fish.

We think with this tool, the work for fillet the fish will be easy without you have an ability like chef and it will make the simpler action than you use only the knife.

 

The mechanism of this tools is firstly we will clip the head of the fish in the clipper to stack the fish in that place. After that, the double knife will stick the fish to get the meat of fish. The distance of double knife is same like the thickness of the bone. Then, the knife will move horizontally until the tail of fish. The last action is we will cut off the fish under the head until the head and the body of fish will separate.

It’s our idea to design the tools for the agriculture (aquaculture). It’s a honor for us if we get the comment to complete the design until make a prototype for this design.

Thanks for the attention.

LET’S TO DESIGN!

notes : click the picture if you want to look it larger and full of the picture.

USAHATANI BAWANG MERAH PADA LAHAN KERING BERSUMBER PENGAIRAN SUMUR POMPA

Studi Kasus Desa Tawali Kabupaten Bima

M. Zairin, Sri Hastuti , I. Basuki ,  dan  Awaludin Hipi

Balai Pengkajian Teknologi Pertanian  (BPTP) NTB

ABSTRAK

Nusa Tenggara Barat (NTB) termasuk daerah kering yang didominasi oleh tipe iklim D4  dan E3    dengan luas wilayah mencapai 20.153.150 ha. Dari luas tersebut lahan sawah seluas ± 220.000 ha (9,33%), lahan kering 1.814.340 ha (84,19%)  yang berpotensi untuk tanaman pangan seluas 330.069 ha, yang dimanfaatkan baru mencapai (6,32%) diantaranya bawang merah seluas 4.927 ha di Bima dengan produktivitas 9,71 t/ha. Bawang merah diusahakan baik di lahan kering (tegal) maupun di lahan sawah.  Pada sebagian daerah sentra produksi bawang merah seperti di desa Tawali kecamatan Wera  kabupaten Bima penanaman pada lahan tegal dimulai menjelang berakhir musim hujan (Maret), yang berlangsung  2- 3 kali tanam selama musim kering. Untuk mengairi tanaman petani menggunakan pompa untuk memompa air baik dari sumur dangkal maupun dari kali.  Usahatani bawang merah pada lahan kering  dengan menggunakan pompa air di Kabupaten Bima, bertujuan untuk mengetahui kelayakan usahatani bawang merah pada lahan kering yang menggunakan pompa air, dan untuk mengetahui kendala  yang dihadapi petani dalam usahatani bawang merah pada lahan kering.  Studi ini dilaksanakan di desa Tawali Kecamatan Wera kabupaten Bima pada bulan Juli 2003, sebagai salah satu desa sentra produksi bawang merah yang memanfaatkan lahan tegal untuk budidaya bawang merah dengan menggunakan pompa air untuk mengairi bawang merah. Jumlah responden sebanyak 15 orang yang diambil secara acak sebagai sampel untuk diwawancarai. Hasil studi menunjukkan bahwa rata-rata produksi yang dicapai oleh petani 6,94 t/ha umbi kering, keuntungan yang diperoleh adalah sebesar  Rp 16.432.010 per hektar dengan B/C ratio 1,92.  Hasil yang diperoleh petani ini lebih rendah dibandingkan dengan hasil kajian sebelumnya pada lahan kering yang menggunakan varietas Fhilipina mencapai  hasil sebesar 14,77  t/ha umbi kering. Rendahnya produksi yang dicapai oleh petani karena menggunakan varietas Bima yang potensi hasilnya lebih rendah dari Fhilipina. Pemanfaatan lahan kering untuk penanaman bawang merah dengan menggunakan pompa air layak untuk dikembangkan dalam meningkatkan pendapatan petani dan PAD.

Kata kunci : usahatani, bawang merah, studi kasus, kabupaten Bima

PENDAHULUAN

Nusa Tenggara Barat (NTB) termasuk daerah kering yang didominasi oleh tipe iklim D4 (4 bulan basah) dan E3 (3 bulan basah) dan bulan kering lebih dari pada 6 bulan (Oldeman, 1980). NTB terdiri dari 9 kabupaten dan Kodya mempunyai luas wilayah mencapai 20.153.150 ha, dengan luas  lahan sawah 200.975 ha (9,33%), lahan kering 1.814.340 ha (84,19%)  yang berpotensi untuk tanaman pangan seluas 330.069 ha, yang dimanfaatkan baru mencapai (6,32%) diantaranya ditanami bawang merah.

Luas areal panen bawang  merah di NTB selama 5 tahun (1997-2001) rata-rata  mencapai 8.544 ha dengan produktivitas rata-rata 5,0 t/ha, diantaranya  4.927 ha terdapat di kabupaten Bima dengan produksi rata-rata sebanyak  9,71 t/ha (BPS NTB, 2002). Hasil kajian (Zairin, dkk,2000) pada uji adaptasi varietas unggul bawang merah di desa Pai Kecamatan Wera Kabupaten Bima MK.I 2000 masing-masing varietas Filipina  memberikan hasil sebesar 15,17 t/ha,  Ampenan 12,01 t/ha,  Sumenep 12,09 t/ha,  Bauji 13, 18 t/ha dan varietas Bima sebagai pembanding 10,8 t/ha dengan rata-rata produksi 12,90 t/ha umbi kering.   Sedangkan hasil penelitian lain di Jawa diperoleh  produksi bawang merah sekitar 10-12 t/ha ( Kusumo dan Soenarjono, 1992).  Rendahnya produksi yang dicapai oleh petani karena teknologi budi daya yang diterapkan masih sederhana tidak menggunakan bedengan pada musim hujan, jarak tanam tidak teratur, jenis dan takaran pupuk juga tidak sesuai dengan kebutuhan tanaman, pengendalian hama dan penyakit belum mengikuti PHT,  mutu bibit yang dipakai relatif rendah serta kesuburan tanah yang rendah.

Untuk meningkatkan intensitas tanam,  produksi dan pendapatan petani pada lahan kering/tegal,  Pemda NTB sejak tahun 1981 s/d 1998 sudah membuat sumur P2AT sebanyak 327 sumur pompa terpasang dengan luas areal yang mampu diari sebanyak 4.952,82 ha dengan melibatkan 327 kelompok tani pemakai air (P3A) sebanyak 2.039 jiwa (Kanwil PU NTB, 1998). Namun masih banyak lahan kering yang belum mendapatkan pengairan dari sumur P2AT karena keterbatasan kemampuan Pemda.  Dalam meningkatkan pemanfaatan lahan kering petani membentuk kelompok dalam menyediakan sarana penunjang berupa pompa air secara swadaya untuk mengairi bawang merah.

Bawang merah merupakan komoditas utama yang memegang  peranan penting dalam perekonomian rakyat.  Bawang merah diusahakan baik di lahan tegalan maupun di lahan sawah,  pada sebagian daerah sentra produksi bawang merah seperti kabupaten Bima penanaman pada lahan tegalan dimulai menjelang berakhir musim hujan (Maret), yang berlangsung sampai 2- 3  kali tanam berturut-turut selama musim kering. Untuk mengairi tanaman, petani menggunakan air sumur dangkal  dengan kedalaman 3-7 m yang di pompa dengan mesin pompa air.

Pengembangan tanaman sayuran dataran rendah yang bernilai ekonomi tinggi seperti bawang merah  dapat membantu usaha diversifikasi usaha tani dalam rangka memantapkan swasemda pangan.  Dengan adanya bawang merah yang bernilai ekonomi tinggi berumur relatif pendek, maka petani mempunyai alternatif lebih banyak untuk memilih komoditi yang sesuai dengan permintaan pasar.  Permintaan pasar akan komoditi bawang merah untuk kebutuhan rumah tangga dan industri pengolahan bahan makanan sekarang semakin meningkat sejalan dengan bertambahnya jumlah penduduk yang menggunakan bawang merah sebagai bumbu penyedap makanan sehari-hari (Duriat.,  1996, Suwandi, 1996,  Rismunandar, 1989).  Komoditi bawang merah Bima NTB diantar pulaukan seperti ke pulau Jawa, Bali, Sulawasi Selatan, Sulawesi Utara, Sulawesi Tenggara dan Kalimantan serta NTT, Maluku dan Papua.  Oleh karena itu Pemda  NTB sekarang sedang gencar-gencarnya mencari dan menggali potensi daerah terutama disektor pertanian pada umumnya untuk menghadapi otonomi daerah guna meningkatkan produksi, pendapatan petani, serta meningkatkan penerimaan asli daerah (PAD).

Studi ini bertujuan untuk mengetahui kelayakan usahatani bawang merah pada lahan kering yang menggunakan pompa air untuk mengairi tanaman, dan untuk mengetahui kendala usahatani bawang merah di lahan kering pada saat musim hujan.

METODA PENELITIAN

Lokasi dan Cara Pengumpulan Data

Survei dilaksanakan pada bulan Juli 2003 di Desa Tawali kecamatan Wera kabupaten Bima, yang merupakan salah satu wilayah sentra produksi bawang merah pada lahan kering/tegalan milik petani yang biasa menanam bawang merah. Data yang dikumpulkan terdiri dari data sekunder dan data primer

  1. Pengumpulan data Sekunder:

Pengumpulan data sekunder dimaksudkan untuk mendapatakan gambaran umum penggunaan dan data yang berkaitan dengan usahatani bawang merah. Jenis data yang dikumpulkan  meliputi harga input, harga output, produksi, luas panen dan produktivitas.

  1. Pengumpulan Data Primer

Data primer dikumpulkan dengan cara wawancara pada petani responden. Penentuan petani responden dilakukan dengan cara purposive sampling, yaitu mengambil 15 petani sampel yang berusahatani berbasis komoditas unggulan (bawang merah), dengan ketentuan terdiri dari kelas penguasaan lahan untuk petani yang mengusahakan tanaman (lahan sempit, sedang dan luas). Lokasi petani sampel ditentukan berdasarkan lokasi sentra komoditas unggulan tersebut di suatu kabupaten/kota. Dari kabupaten/kota diambil lokasi kecamatan sentra komoditas unggulan, dan dari kecamatan diambil petani yang berdomisili di desa sentra komoditas unggulan (bawang merah).

Data ekonomi yang diamati adalah: penggunaan sarana produksi dan tenaga kerja, sedangkan data agronomi: umur panen, produksi dan dan data pendukung seperti serangan hama penyakit.

Untuk mengetahui kelayakan ekonomi teknologi  budidaya bawang merah yang dilaksanakan oleh petani pada lahan kering yang menggunakan pompa air menggunakan formula sebagai berikut  (Anonim, 1988):

RAVC        =  Gross Income – TVC

B/C  ratio  = RAVC/TVC

Dimana: Gross Income   = Nilai Produksi ; TVC    = Total variebel cost atau total biaya berubah

RAVC      = Keuntungan biaya berubah

HASIL DAN PEMBAHASAN

Teknologi usaha tani bawang merah yang diterapkan oleh petani pada lahan kering pada musim hujan (MH) disajikan pada Tabel 1

Tabel. 1   Komponnen paket teknologi usaha tani yang diterapkan oleh petani di desa Tawalii Wera Bima, MH 2002/2003

No

Komponen teknologi

Cara petani

1

Varietas Bima

2

Pengolahan tanah 2 kali Traktor

3

Jarak tanam tidak teratur

4

Jenis& takaran pupuk- Urea (kg/ha)- pupuk daun (sampoerna D)- pupuk buah (sampoerna B) 150-200 kg/ha40 bks40 bks

5

Bedengan tanpa bedengan

6

- Penyiraman- Pengairan dengan alat lokal (pagi atau sore)di leb (kebiasaan petani)

7

Pengendalian Gulma- herbisida sebelum olah tanah- herbisida (7 hst)-  penyiangan tangan Polaris : 3 lt/haRonstar : 1l/ha3 kali  (10, 25, 40 hst)

8

Pengendalian hapen Pestisida (kebiasaan petani)

9

Panen 56 – 60 hst

10

Prosessing dijemur 5-7 hari dan di ikat

Pada MH urea diberikan 3 kali yakni saat tanam, 15 dan 30 hst dengan cara menabur pada lubang tanam kemudian disiram  dengan air supaya tercampur merata dengan tanah dan setelah air meresap ketanah bibit langsung ditanam.

Pemeliharaan tanaman berupa penyiraman pada MH dengan alat penyiraman lokal Bima dilakukan sesaat menjelang tanam untuk memperlunak tanah (lubang tanam) gunakan mempermudah penanaman, selanjutnya tergantung keadaan  curah hujan dan kelembaban tanah, bisa sekali sehari yang dilakukan pada pagi atau sore hari atau berselang. Pada umur 4 mst,  mulai diberikan air pengairan secara leb, yang dipompa dari sumur dangkal yang ada pada lahan petani sekali dalam 5-7 hari (4-5) kali pengairan)/musim tanam  yang di selingi dengan penyiraman  sampai saat panen.  Untuk mengendalikan pertumbuhan gulma terutama yang berdaun lebar (krokot) disemprot dengan herbisida Ronstar 1,0 l/ha pada umur 7 hst.  Penyiangan dilakukan secara manual sekaligus dengan pengemburan tanah/pembumbunan sebanyak 3 kali yakni umur 10, 25 dan 40 hst. Pengendalian hama pemakan daun Spodoptera exygua dilakukan dengan Arrivo atau Bestox 1-2 l/ha.  Untuk penyakit  mati pucuk oleh cendawan Phytopthora porri dan busuk umbi Botrytis allii dengan Dithane M-45 dengan takaran 0,5 – 1,0 kg/ha

Pemungutan hasil/panen dilakukan setelah 80 % populasi batangnya lemas, kira-kira umur  70-90 hari (tergantung varietasnya) dengan cara mencabut (Kusumo dan  Sunarjono,1992).  Namun varietas Bima dapat dipanen pada umur 55-60 hari (Zairin, dkk, 2000). Prosessing bawang merah yang telah dipanen dihamparkan diatas tanah dan dikeringkan  selama ± 7 hari, setelah daun kering kemudian diikat dan dibalik untuk mengeringkan umbi sampai 5 hari.

Penggunaan bibit varietas Bima pada MH. berkisar 1.150 kg – 1.200 kg/ha umbi kering. Tingginya jumlah bibit yang digunakan karena petani tidak menggunakan jarak tanam yang teratur sehingga membutuhkan benih yang lebih banyak.  Biaya saprodi terutama bibit merupakan komponen biaya yang paling banyak dikeluarkan oleh petani yakni berkisar 63,56 dari total biaya.   Bibit yang ditanam oleh petani pada lahan kering/tegal pada MH adalah varietas Bima berasal dari hasil panen sendiri di lahan sawah  pada MK. II (benih antar lapang) yang bermutu rendah.

Kegiatan yang banyak menggunakan tenaga kerja laki adalah penyiraman dengan alat lokal Bima karena dilakukan hampir setiap hari sampai umur 3 minggu dan pada umunya petani menggunakan tenaga kerja sendiri dalam keluarga, kemudian diikuti pemberian air secara leb 5-7 hari sekali dan diselingi dengan penyiraman lagi sampai panen.  Pada kegiatan usahatani bawang merah jenis pekerjaan yang banyak dilakukan oleh tenaga kerja laki adalah pengolahan tanah, pembuatan bedengan, penyiraman dan penyemprotan hama penyakit, dan hanya sebagian kecil pada jenis pekerjaan panen dan prosessing.   Sedangkan tenaga kerja perempuan dominan pada kegiatan pemotongan umbi, tanam, penyiangan, panen dan prosessing.  Jadi pada usahatani bawang merah kehadiran tenaga kerja perempuan sangat memegang peranan penting, karena kegiatannya didominasi oleh tenaga kerja perempuan yang mencapai  50 – 60%,  sedangkan keterlibatan tenaga kerja laki  hanya 40 – 50% dari total kegiatan yang dilaksanakan.

Biaya sarana produksi yang paling tinggi yang digunakan oleh petani adalah bibit bawang merah yang berkisar  63,56% (Tabel 2) dari total  biaya yang dikeluarkan oleh petani selama  proses produksi, sedangkan biaya yang lain termasuk pupuk dan pestisida hanya sekitar 36,44%.

Produksi yang dicapai oleh 15 petani rata-rata sebanyak 6,94 t/ha umbi kering, hasil yang dicapai ini masih dapat ditingkatkan apabila tidak ada serangan penyakit busuk umbi Botrytis allii yang  mencapai 5,0 %,  penyakit  mati  pucuk  Phytopthora porri (25%), dan  adanya umbi yang busuk akibat adanya kelembaban yang tinggi karena hujan terus menerus turun dan tidak menggunakan bedengan. Hasil yang dicapai masih rendah  dibandingkan dengan hasil uji adaptasi varietas  Filipina pada tahun 2000 yang dilakukan pada lokasi yang sama  yakni 15.170 kg/ha umbi kering (Zairin, dkk., 2000).

Produksi yang dicapai oleh petani menggunakan varietas Bima rata-rata sebanyak 6,94 t/ha umbi kering.  Rendahnya produksi yang dicapai oleh petani karena teknologi budi daya yang belum baik, yakni jarak tanam yang tidak teratur, takaran dan jenis pemupukan yang belum memadai, pengendalian hama dan penyakit yang tidak optimal, kesuburan lahan yang rendah.     Di samping itu adanya serangan hama ulat pemakan daun dan penyakit busuk umbi, keterbatasan bibit bermutu merupakan faktor–faktor penyebab masih rendahnya hasil yang dicapai.

Untuk mendukung dan meningkatkan produksi bawang merah diperlukan paket teknologi budi daya yang sesuai dengan kondisi setempat (spesifik lokasi).  Peningkatan produksi bawang merah dapat dilakukan dengan menambahkan bahan organik ke dalam tanah, diantaranya dengan menggunakan kompos, karena kompos mempunyai 2 fungsi yakni (1) sebagai bahan pembenah tanah yang berfungsi memperbaiki struktur tanah terutama tanah kering dan ladang, (2) memperbaiki sifat kimia yang berfungsi mempertinggi kemampuan penukaran kation (KPK)  baik pada tanah ladang maupun tanah sawah (Santoso, B.H., 1998).  Keuntungan penggunaan kompos adalah mampu mengembalikan kesuburan tanah, mempercepat dan mempermudah penyerapan unsur N.  Dalam budi daya bawang merah pada MH penggunaan bedengan dan saluran drainase dapat mencegah terjadinya erosi, mencegah terjadinya genangan air pada tanaman sehingga mengurangi kelembaban yang tinggi  akibat air hujan yang tergenang dan dapat  menyebabkan umbi bawang mengalami pembusukan.

Pasar dan Pemasaran  Hasil

Pasar memegang peranan penting dalam roda perekonomian, karena  pasar merupakan tempat pertemuan antara pembeli dengan produser (petani) yang menjual hasil/produksinya guna mendapatkan uang secara tunai.  Pada lokasi survei pasar desa sudah ada,  pedagang pengecer pada  pasar  tersebut tidak membeli/menjual  bawang merah dalam volume yang banyak, terbatas untuk kebutuhan sehari-hari.   Pasar yang bisa menampung hasil satu-satunya adalah pasar induk di kabupaten, tetapi dalam jumlah yang terbatas  karena konsumen sedikit.  Hal ini sangat menyulitkan petani dalam pemasaran dan menentukan harga yang  layak, sehingga harga banyak ditentukan oleh tengkulak yang datang langsung ke lokasi.

Alur pemasaran hasil produksi bawang merah  dari petani di Bima yakni: a).  petani  ® pedagang pengumpul   ®  pedagang besar    ®   diantar pulaukan (Lombok,  Bali, Kalimantan, Sulawesi Selatan, Maluku). b). Petani ® pedagang pengumpul ® diantar pulaukan (Lombok,  Bali, Kalimantan, Sulawesi selatan, Maluku)  c). Petani ® diantar pulaukan (Lombok,  Bali, Kalimantan, Sulawesi selatan, Maluku). Pada umumnya petani banyak menjual hasil produksinya langsung di lahannya kepada pedagang pengumpul dengan harga yang rendah dari harga produksi. Hal ini terpaksa dilakukan oleh petani karena kalau menunggu harga yang baik memerlukan biaya angkut  dari lokasi ke rumah petani yang cukup mahal.

Untuk mengatasi masalah ini, petani harus bersatu untuk membentuk kelompok tani  yang membantu dalam kebutuhan saprodi dan pemasaran hasil melalui koperasi desa.  Pemda harus turun tangan untuk  membentuk koperasi tani yang dapat menampung hasil dari petani dengan harga yang layak, sehingga tidak mudah dipermainkan oleh tengkulak.  Dengan demikian petani bisa menikmati harga yang layak dari usahataninya sehingga pendapatan petani dan keluarganya bisa meningkat.

Kelompok Tani

Kelompok tani dibentuk dengan tujuan untuk memudahkan pembinaan oleh instansi terkait, memudahkan koordinasi diantara petani yang berhubungan dengan kebutuhan usaha taninya baik saprodi, teknologi budidaya serta pemasaran hasil.  Dalam hal pemasaran hasil kelompok tani tidak mengkoordinasikan sesama anggotanya  tentang kesepakatan pemasaran hasil, dijual kepada siapa, berapa harga jual yang layak, hal ini yang tidak disadari oleh petani sehingga mereka mudah dipermainkan oleh pedagang dalam menentukan nilai jual.   Umumnya petani jarang menyimpan hasil panen untuk menunggu harga yang baik/layak, tetapi segera menjual hasil secara sendiri-sendiri setelah diprosessing kepada pedagang pengumpul dengan harga yang telah ditentukan oleh tengkulak/pedagang pengumpul berdasarkan mutu hasil (ukuran) umbi yakni besar dan kecilnya.

Harga output dan input

Dalam perkembangan terakhir pada usaha tani bawang merah   adalah adanya harga pasar yang naik turun.   Pada saat panen raya, harga bawang merah akan jatuh di bawah nilai/biaya produksi.  Biasanya keadaan ini terjadi pada MK,  sedangkan pada MH atau panen sedikit harga bawang merah tinggi bahkan melebihi harga bawang putih.   Pada pengkajian ini harga yang diterima oleh petani  pada MH sebesar RP 3.600/kg lebih tinggi dibandingkan dengan harga jual langsung  pada MK di lokasi/lahan petani  yang hanya sebesar Rp 3.000/kg.  Perbedaan harga ini cukup besar  yakni mencapai 16,66 %.  Hal inilah yang mendorong petani untuk menanam bawang merah pada MH, disamping harganya tinggi, produk yang dihasilkan oleh petani berapapun jumlahnya tetap laku dan direbut oleh pedagang pengumpul.

Harga output yang diterima petani maupun harga input yang dibayar oleh petani sangat menentukan besarnya tingkat keuntungan usahatani. Secara umum selama dua tahun pengamatan, terjadi kecenderungan penurunan harga komoditi pertanian, sedangkan harga sarana produksi dan upah pertanian cenderung meningkat. Harga rata-rata bawang merah yang diterima oleh petani di NTB pada tahun 2001 adalah Rp 931.250/kw, dan menurun tahun 2002 Rp 552.056/kw (-40,72%) (BPS NTB, 2003).

Produksi

Gambaran luas panen, produksi dan produktivitas bawang merah di Propinsi Nusa Tenggara Barat tahun 2002 dapat dilihat pada Tabel 2. Luas panen bawang merah pada tahun 2002 sebesar 15.484 Ha dengan produksi 58.750 Ton dengan rata-rata produksi 37,94 kw/Ha. Sentra produksi bawang merah berada di Kabupaten Bima dan Kabupaten Lombok Timur, sedangkan luas panen dan produksi tertinggi di Kabupaten Bima.

Sebagian besar areal bawang merah terdapat di Kabupaten Bima (67,11 %), dengan produksi mencapai 81,43 persen dari total produksi bawang merah di Nusa Tenggara Barat. Selain sebagai daerah penghasil bawang merah terbesar, produktivitas bawang merah di Kabupaten Bima lebih tinggi dibandingkan dengan kabupaten lain (97,10 kw/ha).

 

 

Tabel 2. Luas Panen, Produksi dan Produktivitas Bawang Merah menurut Kabupaten di Propinsi Nusa Tenggara Barat, 2002

Kabupaten/Kota

Luas Panen

(Ha)

Produksi

(Ton)

Produktivitas

(Kw/Ha)

  Lombok Barat

574

2.573

44,83

  Lombok Tengah

13

34

26,15

  Lombok Timur

1392

6.476

46,52

  Sumbawa

353

1.086

30,76

  Dompu

83

740

89,16

  Bima

4927

47.842

97,10

Total NTB

15484

58750

80,02

Sumber : BPS Propinsi NTB, 2002

Tingkat keuntungan usahatani

Analisis keuntungan usahatani dalam penelitian ini dilakukan terhadap satu persil dominan (terluas) yang diusahakan petani untuk bawang merah. Pemilihan petani contoh dilakukan di salah satu sentra produksi komoditi yang bersangkutan sehingga diharapkan dapat menggambarkan kondisi usahatani bawang merah.

Pengambilan contoh petani bawang merah dilakukan di Desa Tawali, Kecamatan Wera, Kabupaten Bima. Penaman bawang merah di lokasi contoh dilaksanakan tiga kali setahun yaitu pada MK I, MK II dan MH. Rata-rata luas tanam bawang merah pada MH 2002/2003 yang diusahakan petani berkisar antara 0,28 – 0,38 ha.

Pada komoditi bawang merah juga berlaku hukum ekonomi dimana pada saat suplai tinggi maka harga penjualan akan turun. Sehingga produksi yang tinggi pada MK II 2002 tidak diikuti dengan harga yang tinggi pula. Sebaliknya pada MH 2002/2003 pada saat produksi relatif kurang, harga jual bawang merah menjadi  tinggi. Kisaran harga yang diterima petani bawang merah antara Rp 2.812 (MK) sampai dengan Rp 3.600 (MH.2002/2003).

Tabel 3. Analisa Usahatani Bawang Merah per hektar di Tawali Wera Kabupaten Bima, MH. 2002/2003

Uraian

Fisik

Faktor share dari penerimaan (%)

Faktor share dari biaya saprodi (%)

Produksi (t/ha)

6,94

Harga (Rp/kg)

3.600

Penerimaan (Rp/ha)

24.984.000

100,00

Biaya (Rp/ha)

100,00

 a. Sarana produksi (Rp)
  • Bibit

5.435.323

21,76

63,56

  • Pupuk (Rp)

204.975

0,82

2,40

  • Obat-obatan (Rp)

346.517

1,39

4,05

 b. Tenaga kerja (Rp)

2.265.174

9,07

26,49

 c.  Biaya lain (pompa air)

300.000

1,20

3,51

Total (Rp)

8.551.990

34,24

 Keuntungan (Rp)

16.432.010

 B/C ratio                    1,92

Sumber : Data primer yang diolah

Analisa usahatani bawang merah (Tabel 3) pada MH. di Tawali, menunjukkan bahwa struktur biaya dan penerimaan per hektar dapat dikatakan bervariasi pada setiap musimnya, biaya usahatani bawang merah pada lahan kering pada saat musim hujan (MH) sekitar 34,24 persen dari total penerimaan usahatani.  Sedangkan biaya saprodi yang paling banyak dalam usahatani bawang merah adalah bibit yang mencapai 63,56% dari total biaya saprodi, menyusul biaya tenaga kerja sebanyak 36,49%.  Biaya pompa air untuk MH hanya 3,51% dan biaya ini lebih tinggi lagi apabila ditanam pada MK.I dan MK.II karena sudah tidak ada lagi hujan, sehingga memerlukan penyiraman tiap hari dan pengairan secara leb (5-6 kali) sampai panen yang memerlukan tenaga kerja dan biaya bahan bakar mesin pompa air yang mencapai 14% (Zairin, dkk, 2001).  Dengan struktur biaya seperti di atas, usahatani bawang merah di Kabupaten Bima pada musim hujan (MH) sangat menguntungkan. Hal ini ditunjukkan oleh besarnya nilai keuntungan usahatani yang diterima sebesar Rp 16.432.010 per hektar dengan B/C ratio 1,92.

KESIMPULAN  DAN  SARAN

Dari hasil sturdi ini dapat disimpulkan bahwa:

  1. Usahatani bawang merah pada lahan kering yang menggunakan pompa air di kabupaten Bima memberikan keuntungan sebesar  Rp 16.432.010/ha dengan B/C ratio 1,92.
  2. Masalah usahatani bawang merah adalah adanya serangan hama dan penyakit busuk umbi dan mati pucuk dan ketersedian bibit yang bermutu  (varietas Filipina) masih  kurang, sehingga petani menggunakan bibit lokal antar lapang yang bermutu rendah.
  3. Penggunaan bedengan dan pembuatan saluran drainase pada MH. sangat dianjur kan karena dapat mencegah terjadinya pembusukan umbi oleh air hujan yang tergenang dan penyakit busuk umbi (Botritiis allii).
  4. Penggunaan pupuk kompos pada lahan kering dapat meningkatkan produksi bawang merah
  5. Perlu adanya penangkar bibit bawang merah terutama varietas Filipina untuk mengatasi kekurangan bibit
  6. Permasalah pemasaran merupakan kendala utama peningkatan pendapatan petani. Lemahnya petani dalam kelompok untuk dapat menentukan harga barang membuat petani bawang merah pada barganing posision yang lemah sehingga harga  sangat ditentukan oleh tengkulak. Masalah ini perlu mendapat perhatian yang serius oleh pemerintah.

DAFTAR PUSTAKA

Anonim. 1996. Selayang Pandang Pembangunan Pertanian Tanaman Pangan di Nusa Tenggara Barat. Dinas pertanian Tanaman Pangan Propinsi Dati I NTB, Mataram

Anonim. 1988. Analisa usaha Tani Pola Tanam. Modul  Pelatihan pada Proyek P3NT, Badan Litbang Pertanian. Departemen pertanian.

BPS NTB. 1997. Nusa Tenggara Barat Dalam Angka. Kerjasama Kantor Perwakilan Biro Pusat Statistik Propinsi NTB dengan Kantor Bappeda TK.I. NTB

Duriat,A.S.  1996. Cabai Merah Komoditas Prospektif dan Andalan. Teknologi Produksi Cabai Merah. Bala Penelitian Tanaman Sayuran. Pusat Penelitian dan Pengembangan Hortikultura. Badan Litbang Pertanian.

Kanwil PU NTB.  1998.  Laporan Tahunan Proyek P2AT NTB Tahun 1998

Kusumo S., dan Hendro Sunarjo.  1992.   Petunjuk Bertanam Sayuran.  Cara Bercocok Tanam Bawang Merah. Proyek P3NT Badan Litbang Pertanian. Departemen Pertanian.  hal 51-55.

Oldeman, L.R.,Irsal Las, and muladi.  1980. The Agroklimat map of  Kalimantan, Maluku, Irian Jaya, and Bali, West and East Nusa Tenggara. Central Research Institute for Agricultura Bogor, Indonesia.

Permadi A.H.  2000.  Beberapa Hasil Penelitian Bawang Merah Balai Penelitian Tanaman Sayuran. Balai Penelitian Tanaman Sayuran Lembang. Badan litbang Pertanian

Putrasemedjo, S.  1997. Petunjuk Pelaksanaan Percobaan identifikasi /Multilokasi Varietas Bawang Merah  di Dataran Rendah.  Balai Penelitian Tanaman Sayuran. Litbang Pertanian

Rismunandar.  1989. Membudidayakan 5 Jenis Bawang Merah . Cetakan Kedua. Penerbit Sinar Baru Bandung 1989.

Santoso, B.H. 1998.  Pupuk Kompos. Teknologi Tepat Guna, Penerbit Kanisius.

Sunarjono H. dan Prasojo Soedomo.  1989. Budidaya Bawang Merah (Allium ascelonicum.L). Cetakan Kedua. Sinar baru Bandung. 1989.

Suwandi.  1996. Teknologi Produksi Cabai Merah.  Balai Penelitian Tanaman Sayuran. Pusat Penelitian dan Pengembangan Hortikultura. Badan Litbang Pertanian

Zairin. M., Irianto Basuki, H. Sembiring dan Jafar Abdulgani.  2000.  Uji Adaptasi Varietas Unggul Bawang Merah Pada  Lahan kering Bersumur di Bima.  Laporan akhir Pengkajian IPPTP Mataram T.a. 2000.

Zairin. M., Irianto Basuki, H.Sembiring dan Jafar Abdulgani.  2001.  Kajian SUT Bawang Merah Pada  Lahan Kering Bersumur di Bima. Prosiding BPTP NTB.2001.

 

IT Applications in Agriculture: Some Developments and Perspectives

Friedrich Kuhlmann
Institute of Agricultural and Food Systems Management
Justus-Liebig-University Giessen/Germany
Senckenbergstr. 3, D-35390 Giessen
(Kuhlmann.LBL1@agrar.uni-giessen.de)

The paradise lost of decision-making under certainty
Animals secure their survival and well-being through instinctive actions. Man, however, was driven out of this paradise. Man must consciously establish alternatives of action, predict the consequences of these alternatives, and eventually choose the best course of action for his survival and well-being.
Formally considered, this process is nothing more than the transformation of data into information. Thus, in order to predict consequences of possible actions, we basically need three things, namely
(i) data on environmental variables, relevant to our decision space;
(ii) data on cause and effect relationships within the systems to be employed for our survival and well-being;
(iii) prediction aids as decision support models, which contain the cause and effect relationships and process the data into information.
The expulsion from paradise, however, has had an additional consequence: Whereas animals by means of their instinctive actions always seem to decide under complete certainty, man had to realize that he must decide under uncertainty with incomplete knowledge. He cannot see into the future and only partially grasp the complexity and dynamics of the environment influencing his pondered actions. Hence, the consequences of actions can only be predicted imperfectly and with probabilities. So we need
(iv) knowledge about the “stochastics” connected to the data, as well as to the generated information.

How and where does IT help?
We are convinced that IT helps us in mastering our very existence. Otherwise, we would not tackle corresponding research problems. There are basically three areas of investigation – analogous to production systems for real goods – in which substantial progress has already been made by IT applications, and may certainly be expected in the future.

First, IT supports the production process, i. e. by generating information output from data input by means of models.
For instance, the calculus of differential equations and their particular suitability for the design of dynamic prediction models has been well known for centuries, i. e. since the works of LEIBNIZ, NEWTON, LAGRANGE and EULER, to name just a few. Nevertheless, these systems have only been used to a very limited extent, because (i) analytical solutions can be calculated only for rather simple systems and (ii) manually derived numerical solutions, although possible, are prohibitively time consuming. Hence, alternative calculations with various data sets, in order to explore the decision space, and to conduct sensitivity analysis, were beyond the scope of decision makers. Not before the advent of the computer, could the enormous utility of differential equations for dynamic simulation models unfold.
Also, the basics of (statistical) decision theory have been well-known for centuries, i. e. since BERNOULLI and BAYES. Nevertheless, the theory has – except for simple classroom examples – hardly been used for practical decision support, because analytical solutions were limited to simple problems and manual numerical calculations were too time consuming in this case as well. Again, the usefulness of this quantitative methodology could only unfold when powerful computers became available.
Of course, the ability of computers to perform fast calculations has also triggered the development of new classes of quantitative models. Numerical mathematics, statistical inference, BOOLEan algebra, iterative optimization, and solving of general equilibrium models, may be mentioned in this context.
Second, IT supports the procurement, i. e. the gathering of data as necessary model input.
The tedious counting, measuring and weighing “by hand” or by analog devices has been replaced to a large extent by electronic sensors and digital data collection systems, such that the data may be fed directly into digital decision support models without additional data handling. Automated on-line management of production processes, employing feedback loops, are feasible, as well as the use of geographical information systems for the analysis and simulation of spatially explicit consequences of agricultural and environmental policy measures.

Third, IT supports the logistics, i. e. the transformation of data and information over space and time.
Telecommunication and data warehousing are the preconditions for distributed data processing, involving many agents, logging into the systems at different locations and at different times. An efficient overall management of complete supply chains becomes feasible, as well as data and information exchange for E-Commerce. Not every dream seems to come true, however: Only a few years ago electronic commerce for agribusiness was greeted very enthusiastically. Today, disillusionment can be detected. Even for agricultural commodities, face to face trading seems to be indispensable.

Trends in model development
I will not go into any details concerning the fields of data procurement and logistics. I do not know enough about these problems. Instead, I will concentrate on models as decision-support systems. What developments took place and what developments can and should be expected in the future? I see the following – of course subjectively identified – five trends:
(1) The path of development changes from the construction of predominantly retrospective to mainly anticipative models.
According to the general phase theorem of decision making, planning should pre-empt control to further subsequent pre-post comparison as an information source for corrective actions and for efficiency improvements. However, real life shows a different picture. All empirical investigations with respect to the use of computer models by farmers reveal that most farmers employ only retrospective models to generate descriptive and – at most – diagnostic information. Anticipative models for generating predictive and prescriptive information are used to a much smaller extent.
The recent demand for a complete documentation of activities within entire supply chains has induced an additional impulse for the development of rather sophisticated retrospective models. Traceability and quality assurance are the relevant catch words here, as well as efficient consumer response and just in time delivery.
Nevertheless, in the future, planning models supporting strategic, as well as tactical management tasks – although somewhat more user friendly than at present – should be developed more intensively. Because (as a reminder): Control without planning is impossible, and planning without control is useless.
(2) The path of development changes from the construction of skeleton models to (domain) knowledge-based models.
Up to now, by far the greatest number of models for firm-related as well as for region-related decision support are designed as so-called skeleton models. Not only data for factor inputs but also those for product outputs must be provided by the model user as exogenous entities. The models do not contain any substantial knowledge-based relationships, be it production functions, production rules or behavioural functions, which would be necessary for endogenous predictions of model outputs, based on exogenous model inputs. The prediction of most of the outputs is left to the user.
Reasons for this less than satisfying situation seem to be that agricultural production systems differ in at least two major phenomena from industrial production systems. Contrary to most industrial production systems, where production devices almost completely consist of man-made systems, biotic systems, such as plants and animals are used in agriculture. In industrial branches complex production systems are combinations of simple elements, whose inner structures are well known as a precondition for the determination of production functions. In agriculture it is just the opposite: First one has to break down the complex biotic systems into their basic elements by means of research, to learn more about their inner structures as a precondition for viable predictions of their behavior in response to exogenous input variables. Applied biologists are becoming more and more successful in this area but not so successful that detailed input-output-relationships may be used for concrete predictive calculations. And, for that matter, employing (statistical) black-box-models for the estimation of production functions, does not help very much. These production functions are only valid for the particular experimental plots and vegetation periods from which the data originate. This way, generalized production functions cannot be derived.
In addition, especially the production system “land plus crop” (in rain-fed agriculture) is – with respect to output quantities and qualities – very substantially determined by non-controllable variables like e. g. solar energy and plant usable water. This, however, means that the decision maker cannot control the production system completely. He can only try to optimally adapt the quantities and qualities of the controllable variables (e. g. nutrient supply, plant protection) to the expected values of the non-controllable variables. This is no trivial task, since the values of the non-controllable variables vary over space and time, without the decision maker being able to allocate and predict their values exactly.
In the past, the construction of proper decision aids for this problem area showed only limited progress. But it may certainly be expected to accelerate in the future, if one considers e. g. the efforts concerning precision agriculture or spatially explicit land-use modelling for entire regions.
(3) The path of development changes from open-loop control models to closed-loop control models.
Facing the uncertainty of expectations connected to the risk of false decisions, model designers – whenever this is feasible – replace open-loop control models with models which make use of feedback loops and operate as closed-loop control models. Through more or less continuous monitoring, the decision maker detects deviations between reference values of outputs and their actual values. He then uses these deviations as a base for corrective decisions during process time, i. e. during the growth period of plants or the growth and lactation periods of animals. The on-line approach in the field of precision agriculture (e. g. nutrient supply according to actual crop state), as well as the efficient coordination of complete supply chains, may be mentioned here as examples.
(4) The path of development changes from the construction of models, which assume perfect information about relevant data during the planning period, to models which take into account incomplete knowledge. In other words: Models which abstract from the complexity and dynamics of the relevant environment are more and more replaced by models which explicitly incorporate these phenomena.
Initially, the “number crunching” capacity of computers was used in batch mode. Later on, this mode was replaced by the interactive approach with repeated runs under different data sets, in order to explore the decision space more comprehensively and to identify sensitive input variables.
Meanwhile though, more and more models incorporate the complexity and dynamics of the relevant environment, on the base of sound decision theory, by means of probability distributions for the non-controllable variables. Investment models for strategic decision making, as well as models for operations control of production processes, may be mentioned here.
(5) The path of development changes from the construction of “point in time” and “point in space” models to models which explicitly incorporate time and space.
The characterization of time and space variant non-controllable variables by means of probability distributions is certainly a substantial progress in the model building area. However, it would be even better, if one were e. g. able to directly allocate the values of the non-controllable variables to particular sites within a land parcel and to particular time spans within a vegetation period. If one knew exactly in advance, which values of the non-controllable variables prevail at which sites and in which time spans, one could control e. g. crops with almost perfect information and thus without any substantial efficiency losses. The estimation of probability distributions would not be necessary anymore. However, for the time being, this will no doubt remain a most desirable but hardly realizable state of the model building art.
Nevertheless, we should try to develop such bio-economic models, which explicitly incorporate the relevant time and space variant non-controllable variables, because of their obvious advantages for agricultural production and – for that matter – for the natural environment. They would, at the same time, help to increase production per land unit and decrease the amount of waste, e. g. by not applying more nutrients than needed by the crops.
By means of a very simple example, I would like to show which challenges and opportunities lie before us in this area of model based decision support.

An example: Determining the optimal nutrient supply for a crop
The following example refers to the task of determining the proper nutrient supply (e. g. nitrogen) for a crop (e. g. wheat) on a parcel of land. In order to predict the optimal nutrient supply, we need a production function (as a yield response function), describing the quantitative relationship between the nutrient supply1) and the attainable yield. Suppose, this relationship can be described by a linear response and plateau function, which is a special case of the LIEBIG yield response function (and, for that matter, a special case of the LEONTIEF function). Such a function is depicted in fig. 1. If A is the attainable maximal yield per ha, limited by a given supply of a non-controllable yield factor on a specific site (e. g. plant usable water), then the attainable yield (y) increases with increasing supplies of the controllable yield factor (x), (e. g. the nutrient nitrogen) until the yield plateau (A) is reached. The plant usable water may in this example simply be the sum of the field capacity (nfk), assumed to be completely saturated at the beginning of the vegetation period, and the precipitation (ns) during the vegetation period.
In order to secure the maximal efficiency, the decision maker would obviously take care of the nutrient supply level x1. If he supplies less than x1, the maximal yield would not be attained, thus wasting yield potential. If he supplies more than x1, nutrients would be wasted, since the yield level is limited by the plant usable water. Thus, we have the yield response function
y = min (A; b ⋅ x)
where b is the output-input-coefficient for the nutrient (x). Obviously in this example b = 40.
In reality, however, the supplies of the non-controllable yield factor plant usable water are variable over space and time. The field capacities may vary from site to site within a land parcel. The precipitation levels vary from vegetation period to vegetation period. Thus, in the simplest case, we may have the situation as depicted in fig. 2. On some sites of the land parcel and in some vegetation periods the attainable maximal yield may only be A1, on other sites and in other years, however, it may be A2. If the decision maker does not know where and when the attainable maximal yield is either A1 or A2, he faces a decision problem under uncertainty: If he supplies only x1 = 1,25 dt/ha of the nutrient, assuming the proper yield response function is yA1, he would on some sites and in some vegetation periods forgo the yield Dy = 40 dt/ha. If, on the other hand, he chooses the nutrient supply level x2 = 2,25 dt/ha, assuming the proper yield function is actually yA1, he would on some sites and in some vegetation periods waste a nutrient supply of Dx = 1,00 dt/ha. So, what strategy for the nutrient supply should the decision maker choose?
In a parcel specific strategy certain amounts of nutrients are uniformly distributed over the entire land parcel. Of course, the amounts may vary from parcel to parcel and from vegetation period to vegetation period. In a site specific strategy the farmer employs precision agriculture equipment, and may fertilize with variable rates at different sites within a land parcel, e. g. according to site specific nutrient requirements of the crops.

Parcel specific nutrient supply strategies
Table 1 shows the decision situation for the simplest case one can think of: One distribution for field capacities on the land parcel and one for precipitation levels per vegetation period with two classes each. If we assume, for further simplification, high and low precipitation respectively in 50% of the years and high and low field capacities respectively on 50% of the land parcels, the maximal attainable yields will fall into four classes with probabilities of 25% each. For the example it is assumed (see head of table 1) that the attainable maximal yields are 40, 60, 80 and 100 dt/ha, respectively.

Assuming a value of b = 40 for the output-input coefficient of the nutrient, and by increasing nutrient supplies (xj), uniformly distributed over the entire parcel, we get the yield matrix for the four yield classes shown in the lower left part of table 1. For each of the four yield classes the attainable yields increase linearily until the yield plateaus put a limit to further increases. The column to the right of the yield matrix shows the expected values for the attainable yield per ha of the parcel. The expected values of the yield, as dependent on increasing nutrient supplies, increase initially with constant and then with diminishing rates, until the expected value of the attainable maximal yield level for the land parcel of 70 dt/ha is attained.
The two far right columns of table 1 show the expected values of the nutrient consumption, and of the wasted nutrients. Starting at supply level x4, the nutrient consumption is less than the supply, because already in yield classes with relatively low maximal yields more nutrients are supplied than can be consumed by the plants due to water shortage. The differences between supply and demand (consumption) are wasted.
Under these circumstances, and taking into account the above mentioned information, a risk neutral decision maker would probably choose the action alternative with the maximal expected value of the gross margin. Thus, given the situation outlined in table 2, he would choose the nutrient supply x9. Although in this case the nutrient supply of 2,50 dt/ha would surpass the nutrient consumption of 1,75 dt/ha by 0,75 dt/ha (see table 1), the decision maker will expect the highest value of the gross margin, in this example amounting to 690,00 €/ha.
Usually, however, decision makers do not have knowledge on the probability distributions of the field capacities and the annual precipitations, or, for that matter, do not bother to acquire this knowledge from standard soil maps and long-term precipitation data. Instead, they use some average yields of the past, which may in fact be the expected value of the attainable maximal yield, as a base for the determination of the parcel-wide uniform nutrient supply. Using this yield, however, also implicates that the decision maker assumes an average level of precipitation and an average level of the field capacity for the entire land parcel.

In our example, the decision maker would take the expected value of the attainable maximum yield of 70,00 dt/ha (see table 1) and then principally decide upon the relevant nutrient supply according to the input-output-relationship depicted in fig. 1. In order to attain the maximal efficiency, the decision maker would secure a nutrient supply level of 1,75 dt/ha. Taking into account the afore mentioned product and nutrient prices of 12,00 €/dt and 60,00 €/dt, respectively, he would predict revenues of 70,00 dt/ha ⋅ 12,00 dt/ha = 840,00 €/ha, nutrient costs of 1,75 dt/ha ⋅ 60,00 €/dt = 105,00 €/ha, and thus, a maximal gross margin of 735,00 €/ha.
Table 2, however, shows that because of the probability distributions for the field capacities and the precipitations, a nutrient supply of x6 = 1,75 dt/ha would only result in an expected value of the gross margin, amounting to 615,00 €/ha which is 120,00 €/ha less than the predicted 735,00 €/ha.
Thus, proper knowledge on the probability distributions of non-controllable yield factors, as well as on suitable decision support aids clearly leads to economic advantages, in this case to an increase of the expected value for the gross margin from 615,00 €/ha to 690,00 €/ha.
Besides, the above described facts may also explain why many farmers fertilize more than the strict calculation of the crop’s consumption would suggest. They may have learned from past experience.
For the above described strategies it is assumed that the decision maker has fertilized the land parcel, in order to provide for the envisioned nutrient supply, before he knows about the actual precipitation level of a particular vegetation period. Furthermore, it was assumed that he provides an invariable nutrient supply from vegetation period to vegetation period.
Actually, one can at least think of one more parcel specific strategy which makes use of available additional information. Instead of securing constant nutrient supply levels over the years, in this case the decision maker would apply variable amounts of fertilizers from vegetation period to vegetation period. Moreover, he would apply the fertilizer in several split doses during the vegetation periods, monitoring the growth states of his crop and employing a closed-loop approach as the vegetation period proceeds. Since the growth states of the crop very much depend on the supply levels of the non-controllable yield factors, in vegetation periods with relatively high total precipitation, the decision maker will apply a relatively high total amount of fertilizers, in order to attain the relatively high maximum yields of those vegetation periods. The opposite would be true for vegetation periods with relatively low total precipitation.
Such a strategy is outlined in table 3. In this table the yield matrix is divided in two parts. The (upper) left part shows the yields, as dependent on increasing nutrient supply levels for the vegetation periods with low total precipitation, the (lower) right part those for the vegetation periods with high total precipitation.
The columns to the right of the yield matrices show the expected values of the yields for both precipitation levels, as well as the nutrient costs and the expected values of the revenues and gross margins. In vegetation periods with low total precipitation, the decision maker should obviously secure a nutrient supply level of x7 = 2,00 dt/ha which translates into a maximal gross margin of 600,00 €/ha. In vegetation periods with high total precipitation, he should secure the nutrient supply level of x9 = 2,50 dt/ha translating into a maximal gross margin of 810,00 €/ha. Since it was assumed earlier that the probabilities for vegetation periods with low and high precipitation are 50% each, the decision maker will on average accomplish a maximal gross margin of 705,00 €/ha As a result, acquiring additional knowledge on the values of non-controllable yield factors, by means of their continuous monitoring, may lead to additional economic advantages, in our example to a further increase of the expected values of the gross margin from 690,00 to 705,00 €/ha.

Site specific nutrient supply strategies
Until now, it was assumed that the decision maker uses parcel specific nutrient supply strategies, which in any case lead to uniform levels of fertilization for the entire parcel, although – as in the last case – they may vary from vegetation period to vegetation period. The whole idea of precision agriculture, however, is site specific fertilization according to expected yield potentials on the various sites of a land parcel.
First we shall look into the case where the decision maker does not only know the probability distribution of the field capacities of his land parcel, but has properly located the actual field capacities of the sites within the parcel. On the other hand, the decision maker does not employ the above described closed-loop approach for the nutrient application. Instead, he assumes average precipitation levels to be relevant. In reality, this is often the actual information base.
In this case, the decision maker would actually assume that on sites with low field capacities the attainable maximal yield will be 50 dt/ha as the weighted average of the attainable maximal yields of 40 and 60 dt/ha for “dry” and “wet” vegetation periods, respectively (refer to head of table 1). On sites with high field capacities he would assume an attainable maximal yield of 90 dt/ha as the weighted average for the attainable maximal yields of 80 and 100 dt/ha for “dry” and “wet” years, respectively.
The decision situation which the decision maker actually assumes to be relevant, is outlined in the upper part of table 4. The output-input-coefficient for the nutrient still being b = 40, the decision maker computes the necessary nutrient supply for the sites, having low and high field capacities, with 1,25 dt/ha and 2,25 dt/ha, respectively. According to the assumption that the sites with low and high field capacities each prevail on 50% of the parcel surface, the weighted average of the nutrient supply level is 1,75 dt/ha. The same reasoning applies to the weighted average of the attainable yields, with an expected value of 70,00 dt/ha. Assuming again the product and factor prices of 12,00 and 60,00 €/dt, respectively, the decision maker computes the expected value of the attainable gross margin to amount to 735,00 €/ha (see upper right part of table 4). Table 4. Site specific nutrient supply: Expected and actually attainable results under the assumption of average precipitation, given site specific perfect information on locations and values of field capacities.
However, contrary to the decision makers assumption, there is no constant average precipitation level over time. Given the probability distribution for the precipitation, the decision maker will only gain an expected value of the gross margin of 675,00 €/ha. The relevant computation is shown in the lower part of table 4. In applying the calculated nutrient supply strategy, the decision maker will only obtain an expected value for the attainable maximal yield of 65,00 instead of 70,00 dt/ha. This is mainly due to the less than sufficient nutrient supply for the sites with high field capacities in vegetation periods with high total precipitation. With a nutrient supply of 2,25 dt/ha he will only produce a yield of 90,00 dt/ha, although the amount of plant usable water would be sufficient for a yield level of 100,00 dt/ha.
As a result, the expected value of the gross margin for this site specific nutrient supply strategy is less than it would be, if the decision maker would only make use of the probability distributions of the field capacities instead of taking them directly into account (675,00 €/ha compared to 690,00 €/ha). The negative difference is even higher when the parcel specific nutrient supply strategy is employed with the closed-loop approach for fertilization (675,00 €/ha compared to 705,00 €/ha).
In other words: Ascertaining only the levels of the field capacities of the different sites of the parcel and not taking into account the variability of the precipitation levels does not lead to economic advantages over parcel specific nutrient supply strategies, taking into account probability distributions for non-controllable yield factors.
Site specific nutrient supply, however, is becoming economically superior, if the decision maker uses site specific knowledge of the field capacities and employs the closed-loop approach for fertilization. Formally, this would be decision making with perfect information.
Relying on the data about the precipitation level, gathered by the monitoring process during the vegetation period, and knowing the field capacities of the different sites of his parcel, the decision maker calculates the relevant nutrient supply levels to be 1,00, 2,00, 1,50, and 2,00 dt/ha for the four possible supply levels of the non-controllable yield factor plant usable water. Under these conditions he will attain an average gross margin of 735,00 €/ha for the parcel. Since, in this case, the a-priori known values for the field capacities and the total precipitation per vegetation period are identical to their a-posteriori values, the decision maker employs a strategy for which expectations and results are identical.
Of course, the attainable value of the gross margin will change from vegetation period to vegetation period, due to the time variant precipitation levels. But in the long run, the decision maker can expect an average gross margin of 735,00 €/ha, provided – of course – the a priori-assumed probabilities for the precipitation levels are identical to their future values, i. e. do not change over time.
In our example, using site as well as time specific information as a base for the derivation of proper nutrient supply strategies, leads to an economic advantage of 45,00 €/ha over the parcel specific strategy, taking into account only the probability distributions for the field capacities and the precipitation levels (735,00 €/ha compared to 690,00 €/ha). The site and time specific nutrient supply strategy has the additional advantage of wasting less nutrients leading to environmental improvements (e. g. lower nitrate concentrations in the ground water).
Beware, however, of jumping on conclusions too soon. In order to apply the site specific supply strategy, the farmer needs to make additional investments to determine the site specific field capacities and in the equipment for site specific fertilization. Only if the additional costs for these investments amount to less than 45,00 €/ha, the site specific nutrient supply strategy is – in strictly economic terms – comparatively advantageous.


Conclusions
In conclusion, the simple example suggests that we should develop decision support systems which properly take into account the complexity and dynamics of the systems environment which comprise subject matter, as well as economic components. In other
words: One major application of IT in agriculture will certainly be the development of knowledge-based, bio-economic models which
(i) will contain appropriate input-output-relationships as generalized production functions,
(ii) will take into account space and time variability by incorporating the relevant, non-controllable yield factors, preferably with their direct values or at least with their probability distributions, and
(iii) will contain biological and technological, as well as economic components, in order to provide effective decision support for the agricultural land users.
Obviously, such models will have to be developed by multi-disciplinary teams, comprising subject matter experts from the fields of agriculture and business management, as well as computer scientists and statisticians. In order to guarantee practical relevance and user friendliness, extension specialists and selected professional farmers should be involved in the model building process as early as possible.
As a rule, the development of a decision support model may be designed as a stepwise and iterative process. A thorough systems analysis and the establishment of a theoretical concept should lead to the prototyping of a first model version. These steps will be followed by tests on research farms and by extension specialists and farmers, as a base for model enlargements and refinements. After several model adaptations and improvements the final product will eventually be ready for the farming community.
While the marketing will typically be conducted by public or private service providers, subject matter research and model development will typically lie in the hands of university groups or specialized institutions for applied research. Since the agricultural sector, being composed of many small business entities is not able to finance these research and development efforts, funds will have to be provided through governmental agencies and research foundations, of course, based on competitive bidding and peer group reviewing of proposals.
With such models we will certainly not regain the paradise lost of decision making under certainty by way of instinctive actions, but we will eventually be able to support decisions which yield results being technically and economically less inefficient than they are usually now.

Dr. Dr. h. c. Friedrich Kuhlmann is Professor of Farm Management at Justus-Liebig-University of Giessen (Germany) since 1973. He studied agricultural sciences, business management and systems science at the universities of Berlin, Giessen and Michigan State. His research interests comprise agricultural production economics, decision theory and land use planning methodology. He teaches courses in agricultural production management and quantitative modeling. In addition, as head of a research farm for farm management he develops – based on longstanding practical experience – IT-based decision support models for agricultural production systems. He has held positions as chairman of the German Association of Agricultural Economists and as a member of the Scientific Council with the German Ministry of Agriculture. He is currently Vice-President of the German Agricultural Society (DLG) which is the major institution for knowledge transfer in the German agricultural and food sector.

Notes :

If you want to download it, you can download from http://scholar.google.com and you can write this subject.. I can’t give it because this post maximum only for300 kb.
Thanks..