Workshop on Systems Analysis: Summary Report (1984)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

PART II INDONESIAN PRESENTATIONS, WITH COMMENTS BY NRC PANELISTS

PRELIMINARY STUDY OF A WASTE DISPOSAL SYSTEM FOR CENTRAL JAKARTA Team for the Solid Waste Management Study,* Directorate for Systems Analyses, Agency for the Assessment and Application of Technology (BPPT) The preliminary study of a waste collection and transportation system and the preliminary design for a waste incineration plant for Central Jakarta described in this report constitute part of the Study of the City Development System, conducted by BPPT's Directorate for Systems Analyses, in cooperation with the Special Government of Jakarta. (Jakarta is designated as a Special Capital Area and divided into five mayoral districts, i.e., Central Jakarta, Northern Jakarta, Eastern Jakarta, Southern Jakarta, and Western Jakarta.) The sanitary landfill, composting, and incineration methods of waste disposal are currently being used in Central Jakarta. Incineration is a solid waste processing method that utilizes the chemical process of oxidation. A BPPT study of waste characteristics in Central Jakarta has revealed that the solid waste in that area meets the minimum require- ments for its use as the basic material for a waste incineration plant (WIP), which will utilize energy recovery equipment. Several countries, unable to apply the sanitary landfill method because of insufficient space, have improved upon WIP technology to the point that it can process more than 30 percent of total waste. In this chapter the technological aspects and requirements of a WIP are reviewed, keeping in mind that a WIP must be able to operate safely, must not disturb the environment, and must be in accord with the econo- mic condition in general. The first section describes waste quantity, quality, collection, and transportation at present, and plans for their future improvement. The second section describes BPPT's study of waste incineration—location determination, investment costs, operational costs, and revenues gained from the sale of electricity to PLN (state- owned electricity corporation). The third and final section compares the advantages and disadvantages of the waste incineration method and the sanitary landfill method. This study was conducted from October 1981 to May 1982 by the Directorate for Systems Analyses, with the assistance of Geopfert, Reimer und Partner, consultants, Hamburg, Federal Republic of Germany. *Sri Bebassari S., team leader; Djoko Herumartono; Endang Trihadiwati; H. B. Henky Sutanto; Lusina Waluyati; M. Ansorudin S.; Made Setiawan; Rachmat Wiranegara; Rahardjo; Siswanto Sewoyo; Sri Rahayu; and Tasmian. - 13 -

- 14 - WASTE GENERATION IN JAKARTA, PARTICULARLY CENTRAL JAKARTA In 1981, waste generation in Jakarta was estimated at 1.2 million tons per year, assuming 2.5 liters (0.5 kg) of waste per capita per day, while that in Central Jakarta for the same year was estimated at 270,000 t/yr, assuming 3 liters (0.6 kg) of waste per capita per day. Future waste generation will increase proportionately to increases in population and people's living standard (see Annexes A and B). The growth rate of waste generation until 1995 is estimated at 4 percent per year, but it is expected to decrease to 3 percent per year for the following decade. The actual figures for waste generation are given below: Jakarta Central Jakarta Year (thousand t/yr) (thousand t/yr) 1981 1,200 270 1985 1,400 300 1995 1,900 400 2005 2,300 470 As discussed and considered in the BPPT study, this waste consists only of domestic solid waste and similar waste from residential areas, offices, shopping centers, market places, and street cleaning. Composition and Characteristics of Solid Waste in Central Jakarta The BPPT study conducted in 1981 found out that 80 percent of solid waste from Central Jakarta consists of organic matter with the following characteristics (see Annex C): Rainy Season Dry Season Water content (percent) 63 58 Ash content (percent) 9 12 Heating value (kJ/kg) 4,500 6,152 The following influences will, however, change the above composition and characteristics of waste:

- 15 - • A change in living standard. The prediction of a rising standard of living implies that more wrappings and packaging material will ultimately end up as domestic waste. In Central Jakarta, the waste from office and administration buildings will be especially high, particularly the amount of paper. • An improved collection and transportation system. The projection of improved collection, for example, by enclosed containers and transportation in covered or closed trucks means that waste will not be exposed to rain. Thus its water content will decrease, especially in the rainy season. Because of these influences, the water content will decrease, the amount of combustible matter will increase, and the density of solid waste will decrease, leading to an increase in the heating value. Such a trend thus has good implications for the basic materials of a waste inciner- ation plant (see Annex D). THE EXISTING SYSTEM OF COLLECTION AND TRANSPORTATION The solid waste of Central Jakarta is currently collected directly from its sources—households, markets, commercial centers, home industry, and street sweeping—where it is usually stored in containers of different types and shapes. In residential areas with a single storage house, domestic waste is stored in concrete boxes, of which only a few have covers. In some areas, the solid wastes are thrown into an open site or into rivers and the surrounding area. In general, solid waste is currently collected and transported by cart from its source to the temporary dumping ground, and then further transported by different types of trucks to the disposal site. In another collection and transportation system, the solid waste is stored in containers of different types and collected by a door-to-door service that uses different types of trucks such as compactors, open trucks, and dump trucks (tipper trucks) to transport the waste to the disposal site. These collection vehicles serve only about 7 percent of the total population of Central Jakarta. A field survey conducted by BPPT in March 1982 found that: • Only an estimated 490 t/day or 56.6 percent of the total 868 t/day of waste generated in Central Jakarta can be trans- ported. The transportation of solid waste is carried out by the Cleaning Division of the Central Jakarta Administration and by the PD Pasar Jaya Corporation which is owned by the regional administration. • A total of 143 trucks were used in March 1982, of which only 15 percent were not older than 5 years, the average service life of such trucks. • Working 8 hours a day, the solid waste fleet averaged only 1.75 trips per day because of road conditions and heavy traffic and a lack of fixed routes from the temporary dumping site to the disposal site.

- 16 - • Because the equipment being used in the existing system of col- lection and transportation is inadequate, both in quantity and quality, the working hour is indirectly prolonged, thus lowering transport capacity. The capacity of waste transport can be increased by adding transport fleet and improving their system of collection and transportation. An Improved Waste Collection and Transportation System An improved waste collection and transportation system is also needed to ensure a continuous supply of waste to a waste incineration plant. To support this effort, six models of waste collection and transporta- tion systems have been developed in the BPPT study. These models are based on established figures to be able to compare data from (more or less) equal sources: 1. One service unit equals one kelurahan (administrative unit) having an average population of 30,000, or approximately 6,000 households, assuming each household contains five persons. 2. Waste production is assumed to be 3 liters (0.6 kg) per person per day, so that the total production of each service unit is 90 m3/day, or 18 t/day (at the density of 0.2 kg/liter). 3. Since 80 percent of waste easily decomposes within a few days, waste and garbage are collected at least twice a week. 4. Waste collection and transportation are carried out 6 working days a week (52 weeks a year, or 312 working days a year). The capacity of transport service required is 21 t/day, or 6,552 t/yr for each service unit. The specification of each model is shown in Annex E. Model I uses the simplest equipment and working method, while Model VI uses more sophisticated equipment and working method. Each model is designed according to three components: collection system, transfer system, and transportation system. The collection systems described in Models I-IV use carts, while Models V and VI use compactor trucks with a door-to-door collection system. The transfer system described in Models I, II, and V is carried out manually; Model III uses a crane; Model IV uses containers; and Models V and VI use compactor trucks. Of the six models, Models IV and V are thought to be most applicable to Central Jakarta, with adequately low specific costs (see Annex F), i.e., Rp. 12,829 (rupiahs) per ton of waste (Model IV) and Rp. 12,412 per ton of waste (Model V). Model IV, however, which uses container trucks, requires land that is not always available for a container transfer station, while Model V, which uses a compactor truck, requires a wide road. The improved waste collection and transportation system will thus consist of a combination of these two models, adapted to the conditions of Central Jakarta. Using Model IV, the cost of waste collection and transportation in Central Jakarta in 1990 will be Rp. 4,448.5 million and Rp. 4,303.7 million for Model V.

- 17 - SOLID WASTE TREATMENT BY INCINERATION (PRELIMINARY DESIGN) The planned waste incineration plant will be equipped with discharging and storage space, a boiler, a slag bunker, a dust electrostatic pre- cipitator, and wastewater treatment. Steam produced by the boiler will be used to operate the turbine which is connected to a power-generating set. Waste with a minimum heating value of 4,500 kJ/kg and a water con- tent of 62 percent must be incinerated to a temperature of at least 800°C to guarantee complete incineration and, at the same time, the elimination of any waste odor that may be emitted with the smoke. Solid waste from Central Jakarta meeting these conditions can reach the required temperature during normal operation without the assistance of additional fuel. If the waste incineration plant is established in 1983 and commences operation in 1985, with 20 years of service life, the plant will be operational up to 2005. Should Central Jakarta's waste production reach 470,000 t/yr in 2005, and given the plant's incinerating capacity (efficiency) of 70 percent, four incinerator units must be built, each with a waste capacity of 20 t/h/unit. In 1985 (first year of operation) Central Jakarta's waste production is estimated at 300,000 t/yr; thus only three incinerator units will be needed that year, requiring 400,000 tons of waste. To meet this demand, waste from outside Central Jakarta can be used. The first three incinerator units will generate 12.42 MW of electrical energy during the rainy season and 17.22 MW during the dry season, with an average annual generating capacity of 96,000 MWh. The incineration plant itself consumes 16,000 MWh of electric energy per year; thus 80,000 MWh of electric energy can be distributed yearly to the PLN network. Another product of incineration is slag, which is estimated at 216 t/day, or 60,000 t/yr (from the first three units). This slag should be removed into a discharging location that will not pollute the environment. It can, however, be used for land reclamation and for building or road material. The amount of flue gas produced by the plant will be 360,000 m /h, and the greatest pollutants are carbon monoxide at 2.648 mg/m and nitrogen oxide at 935 mg/m . By using a chimney 90 m high, the concentration of pollutants will be lower than the acceptable standard applied by the Jakarta Administration. Three alternative locations in Jakarta for the WIP have been discussed with regard to their geographic suitability, city development planning (land use, transportation, PLN network, etc.), available natural resources (processing water, cooling water), and airplane traffic: Cakung, Tanjung Duren, and Kebon Jeruk (see Annexes G.I and G.2). It has been concluded that Cakung is the most optimal location for the establishment of a waste incineration plant.

- 18 - Economic Analysis The operational costs, which consist of fixed costs and variable costs, will change according to the inflation rate. Based on last year's inflation rate, the average increase in cost is calculated at 10 percent per year. Revenues from electric energy sales are also expected to increase in line with the inflation rate. For economic calculation purposes, the average increase in both prices and revenues are estimated at 10 percent per year. The result of the calculation is shown in Annex H which describes operational costs and revenues for several years to come, assuming a sale price for electrical energy of Rp. 28 kWh and an interest rate of 9 percent per year (also see Annexes I and J). According to these calculations, the plant will suffer decreasing deficits from year to year until the deficit reaches zero at the end of the nineteenth working year. City waste incineration plants in other countries receive subsidies from the central government. In the Federal Republic of Germany, for example, the subsidy amounts to 20-40 percent of the total incineration cost. COMPARISON OF A WASTE INCINERATION PLANT AND A SANITARY LANDFILL In the following comparison of the sanitary landfill and incineration methods, it is assumed that the two methods are in service for a period of 20 years. It is also assumed that the waste incineration plant is located in Cakung, while the sanitary landfill is located in Pondok Ranggon. Waste Incineration Comparative Factor Plant Sanitary Landfill • Waste composition Nixed waste Mixed waste • 20-year operation 5,475 working days 6,240 working days period • Amount of waste that 7,884,000 t 8,985,600 t can be disposed of in 20 years • Land area required 3.5 ha 297 ha • Average transport 15 km 21 km distance • Possible pollution -In air -Surface -Slag -Strong odor • Revenues Electrical energy None • Specific cost per Rp. 11,596 Rp. 10,838 ton of waste

- 19 - CONCLUSION It can thus be concluded that the establishment of a waste incineration plant for Central Jakarta is technically feasible. Should the WIP be established, the following matters should be taken into consideration: • The continuous supply of waste demanded by the WIP must be maintained. • Waste characteristics, i.e., heating value and water content, should be maintained to meet the requirements of the WIP. It is understood that the water content of waste in Central Jakarta is 60 percent of the total weight. • The existing system of waste collection and transportation should be improved to both ensure a continuous supply of waste and decrease the water content of the waste. • Of the six models of the collection and transportation system being developed, Model IV (using container trucks) and Model V (using compactor trucks) appear the most applicable. Model IV requires land, which is not always available, for a container transfer station, while Model V requires a wide road. Thus an improved waste collection and transportation system will constitute a combination of these two models. The specific cost of Model IV is Rp. 12,829 per ton of waste, and for Model V, Rp. 12,412 per ton of waste. • Based upon Central Jakarta's projected waste production in 2005, the WIP will consist of four incinerator units, having a waste capacity of 20 t/h/unit, to be constructed in two stages. In the first stage, three incinerator units will be built, capable of incinerating 400,000 tons of waste per year and generating 96,000 MWh of electrical energy per year. • The WIP will require an investment of about Rp. 49,962 million (based on the 1982 price level), with an operational cost of about Rp. 5,496 million per year for a period of 20 years, at an annual interest rate of 9 percent. If electricity is sold to PLN at Rp. 28/kWh, there will be a yearly deficit which will reach zero in the nineteenth year of operation. Thus the cost of incinerating 1 ton of waste will be Rp. 11,596. When compared with the sanitary landfill method, the establishment of a WIP in a densely populated area like Jakarta is more appropriate, both technically and economically. REFERENCES Studi Keterlaksanaan Permasalahan Sampah di DKI Jakarta (P4L-DKI Jakarta 1979). Pilot Project SWIP (Solid Waste Improvement Program for Jakarta 1981).

- 20 - ANNEXES Annex A Waste production increase (cubic meters/year), DKI Jakarta and Central Jakarta Annex B Waste production increases (tons/year), DKI Jakarta and Central Jakarta Annex C Waste composition and characteristics, Central Jakarta, 1981 Annex D Waste specification, DKI Jakarta and Central Jakarta Annex E Technical data in the models of the waste collection and transportation systems Annex F Operational costs of the waste collection and transportation models Annex G.I Evaluation of location based on planning and environmental considerations Annex G.2 Evaluation of location based on economic considerations, and the results of the location evaluation Annex H Increased operational costs of and revenues from waste incineration Annex I Waste incineration plant: investment, operational costs, and revenues Annex J Operational costs of the waste incineration plant over the period of operation

- 21 - ANNEX A WASTE PRODUCTION INCREASE (CUBIC METERS/YEAR), OKI JAKARTA AND CENTRAL JAKARTA in ) Nomxrud aisvM

- 22 - ANNEX B WASTE PRODUCTION INCREASES (TONS/YEAR), DKI JAKARTA AND CENTRAL JAKARTA OIOZ

- 23 - ANNEX C WASTE COMPOSITION AND CHARACTERISTICS, CENTRAL JAKARTA, 1981 SEASON I. WASTE COMPOSITION DRY RAINY AVERAGE 1. Organic waste 76.99% 81.99% 79.49% 2. Paper 7.83% 8.11% 7.97% 3. Wood 4.96% 2.33% 3.65% 4. Textile 2.74% 2.07% 2.40% 5. Rubber /imitation leather 0.40% 0.54% 0.47% 6. Plastics 4.02% 3.32% 3.67% 7. Metal 1.52% 1.22% 1.37% 8. Glasses 0.62% 0.38% 0.50% 9. Others (soil, stone, sand) 0.92% 0.04% 0.48% II. WASTE CHARACTERISTICS 1. Water content 57.51% 62.67% 60.09% 2. Ash content 12.44% 8.73X 10.59% 3. Heat value (kJ/Kg) 6,152.00 4,501.00 5,326.00 SOURCE: BPPT (1981) Brief Study of Waste Composition and Characteristics in Central Jakarta.

- 24 - ANNEX D WASTE SPECIFICATION, DKI JAKARTA AND CENTRAL JAKARTA KEY: O .V CENTOU, JAKARTA DKI JAKARTA

- 25 - ANNEX E TECHNICAL DATA IN THE MODELS OF THE WASTE COLLECTION AND TRANSPORTATION SYSTEMS 1 CM 8 i 1 1 1 P« .*• '( 1 1 1 1 Q » N . * S . S ...4 O <M *-i O O . CO cO O^ O ••O i c u I o i B8 cn eg -i N ^ co fT So CM *n f-i u 0 • 3 -4 I £ i M5 O 1 II I I 4 1 1 I 1 1 S * . cn CM 00 O 3 BOO (Ti . <J • a O t-i - o - !§ •0 = ^ M r- o\ B CM r-l CM .* rH c art H > <T O I-4CMM9W U mifl CM** . . <J . o HI O i-4 c MCM .CMcocMcMcnom & ^ '< M «B '• « M sf 00 a cn M » CM 1 8S 8 2 CO 1 M M H 3 U M ^o »-« c-i \o a «n o M m «n CM . oo O 1-4 « X X CM * - - 03 0 m 01 M • b * O cs r-* - • 00 B •O O .— ' O \£) OO m a X E K CO O CN M r-~ 04 HH jr CO JJ -O CM P4 ,2 H U J5 3 M M cnOO^-^00 .-. 1 101 ^ ^t^r^c*^^*^^ »ito '-H cj ^o to i i »n i o o CM \o ^ CM . ao r— I X.*» g .-4 O i-1,9 ,-' 9 . . . .£> o in g H * a MO -*-*•* § j; j; s s .H CM •— , g. i. o H cnopcnooo ^^ i 101 Ijrt mcnr^cMr*r*-*cnr* i«t o cn^\o4 1imiiu CM «\OCM*CO o 8 o o M - M O CM H! — ' 9 . **«OOIA cn £-4 . o a H u co cn I -» M r» «M n z 04 3 •n -» ss sS CM / H g> / 1 / £ JtfbVX4J4Jb •H CMbVOi ma X cwbBb t) / V«UOJ•H.^V b Q X —« rH HO •* rHX^X « * ai u ai TJ a c j< « '- a. o. -».*j -o o.~>^ — — » -^ -H ^ 3 5 b £ j:oo a — oBbb — -j a ai o -» x 01 o o • • Ji • • w » a ji a a u UO.UW 4-1 / g § 2 § 3 3 .O J / b cn b B 4-i a / .3 S 3 / f H x u a a M 1 3E X b X -H u « 0 -H --i -— cau xovoo U M ov COM Bdauu H awx o aibxa.0. / 01 «4J-HU Q Z 01 3OUBB >• 01 <-i 3 C -" H Ob O-O tr«W .133 u 3 ai o LA >* x a> 4-1 « HM Xbv taca §a FH 5 b • M to u bo.-n .< If fi Ik m oo « O fl MJ « — 1 -H to B > n H 01 T9 bi-IUU M3>« — o. a* 3b«Mi — j<j<«)ti .— < [10* UU >U •4JT3«0ITIO3-O'OOO>>-1— 1 M uvacbbou b, a -H a o a. «l a,^<offl33-H«aiai k|b-H-Hoj«T4b CO 0! C ffl o 3 O COOOObbbb33 jjkioqcauuaau x u9ijcjca.j Z>JJHHOH(X,U. / M / g 3 M H u u^cMcn^mor- H r4cMcn^rm\o H^cMcn«mM>r»ooc7\ Z

- 26 - OPERATIONAL AND COSTS OF THE WASTE COLLECTION TRANSPORTATION MODELS ANNEX F ^>v- MODEL EXISTING BINS, CARTS. TRUCKS ADVANCED BINS-CARTS-SYSTEM BBHMena ADVANCED COMPACTOR DOOR-TO-DOOR TRUCKS TIPPER CRANE CONTAINER DOOR-TO- TRUCK DOOR COST COMPONENT I II III IV V n COLLECTION SYSTEM (Kp/yr) Investment cost - Domestic u>e bin - Cart (gerobak) - Cart bin 10,080,000 2,656,000 17,107.200 10,080,000 1,328,000 8,554,000 10,080,000 1,328,000 8,554,000 10,080,000 20,160,000 1,328,000 — 8,554,000 — 35,280,000 Fixed cost - Workers - Administration 19,440,000 1,944,000 19,440,000 1,944,000 19,440,000 1,944,000 19,440,000 — 1,944,000 — Collection coat (Rp/yr) Specific coit (Rp/ton) 51,227,200 7,819 41,346,000 6,310 41,346,000 6,310 41,346,000 20,160,000 6,310 3,077 35,280,000 5,385 TRANSFER COST (Rp/yr) Investment coat - Lend 9.000,000 9,000,000 9,600,000 11,250,000 - Equipment 3,375.000 — Fixed cost - Maintenance - Worker* 11,208,000 1,121,000 11,208,000 1,121,000 3,750,000 1,920,000 192,000 1,125,000 — 1,800,000 8,400,000 180,000 840,000 4,200,000 420,000 Variable cost. - Power conaraption 1,363,000 Transfer coat (Rp/yr) Specific coat (Rp/ton) 21,329,000 3,255 21,329,000 3,255 28,075,000 4,285 15,480,000 9,240,000 2,363 1,410 4,620.000 705 TRANSPORTATION COST (Rp/yr) Investment coat - Trucka 21.015.000 21,015,000 11,650,000 17,500.000 35,000,000 39,375.000 Fixed coat - Maintenance - Horkera 4,203,000 3,362,000 336,000 841,000 4,203,000 3,362,000 336,000 841,000 2,330,000 1,678,000 168,000 466,000 3,500,000 7,000,000 2,016,000 5,040,000 202,000 504,000 700,000 1,400,000 7,875,000 2,520,000 252,000 1,575,000 - Administration - Inaurance Variable coat - Fuel - Tirea 2,005,000 755,000 2,005,000 755,000 2,005,000 755,000 2,406,000 2,506,000 906,000 472,000 1,504,000 566,000 Tranaportation coat (Rp/yr) Specific coat (Rp/ton) 32,517,000 4,963 32,517,000 4,963 19,052,000 2,908 27,230,000 51,922,000 4,156 7,925 53,667,000 8,191 TOTAL OPERATION COST (Rp/yr) SPECIFIC COST (Rp/ton) 105.037,200 16,037 95,142,000 14,529 88,473,000 13,503 84,056,000 81,322,000 12,829 12.412 93,567,000 14,281 ROTE: Everv operational coat ia calculated ac .000 people).

- 27 - ANNEX G.I EVALUATION OF LOCATION BASED ON PLANNING AND ENVIRONMENTAL CONSIDERATIONS* TANJUNG KEBON ASPECT VALUE CAKUNG DUREN JERUK City management 27 1. Land use 10 30 20 10 2. Surrounding factories that will consume the electricity 4 8 8 8 3. Wind direction in the rainy season 8 16 8 8 4. Others 5 10 10 10 Air pollution before the waste incineration plant opens 1. Dust 2 4 2 2 2. Smog 3 6 6 3 3. Noise 3 6 6 3 Air pollution after the waste incineration plant opens 1. Dust 2 4 4 4 2. Smog 3 6 6 6 3. Noise 3 6 3 3 Hydrology 7 1. Groundwater 4 12 12 8 2. Surface water 1 1 1 1 3. Drinking water 1 3 3 1 4. Waste water 1 2 2 1 Results 100 114 91 68 *The values are based on the team's agreement.

- 28 - ANNEX G .2 EVALUATION OF LOCATION BASED ON ECONOMIC CONSIDERATIONS* ASPECT VALUE CAKUNG TG. DUREN KB. JERUK Costs that are influenced by waste incineration plant locations 15 1. Land cost 2 . Roads 3. Water 4. Electricity network 4 4 4 3 8 12 12 9 4 12 8 6 12 8 Waste incineration plant, installation cost 10 4 6 1. Preparing the land 2 . Building 3. Infrastructure 4 4 12 12 6 3 3 3 3 6 6 6 Transportation cost 25 1. Waste transportation 2. Slag transportation 3. Others 20 2 3 40 6 9 60 2 9 40 2 6 Results 100 106 125 102 *The values are based on the team's agreement. WASTE INCINERATION PLANT, RESULTS OF THE LOCATION EVALUATION ASPECT VALUE CAKUNG TG. DUREN KB. JERUK Planning & environmental Economic 100 100 114 106 91 125 68 102 Results 200 220 216 170

- 29 - ANNEX H INCREASED OPERATIONAL COSTS OF AND REVENUES FROM WASTE INCINERATION (Rp. 106) YEAR 1982 1987 1992 1997 2002 COST COMPONENTS 1 Investment 5 ,496 5 ,496 5 ,496 5 ,496 5 ,496 2 Fixed + variable cost 1 ,383 2 ,227 3 ,587 5 ,777 9 ,304 . Total cost (1+2) 3 6 ,879 7 ,723 9 ,083 11 ,273 14 ,800 4 Revenue* 2 ,240 3 ,607 5 ,810 9 ,357 15 ,070 . Total deficit (3-4) 4 4 5 ,639 ,116 3 ,273 1 ,916 — Specific cost 11 ,596 10 ,290 8 ,182 4 ,790 -•. (Rp./ton waste) *Electricity selling price in 1982 is Rp. 28/Wh. NOTE; The calculation is based on: 1. Interest rate, 9%/year 2. Operation cost increase, 10%/year 3. Increase in electricity selling price, 10%/year

- 30 - ANNEX I WASTE INCINERATION PLANT: INVESTMENT, OPERATIONAL COST, AND REVENUES Interest Rate 6Z/Year 9Z/Year 12Z/Ye*r 1. 2. Work Capacity Elec. Distribution Ton/Year MWh/Year 400,000 80,000 400,000 80,000 400,000 80,000 INVESTMENT - Mechanical - Electrical - Building Rupiah Rupiah Rupiah 24,292,000,000 5,258,000,000 10,757,000,000 24,292,000,000 5,258,000,000 10,757,000,000 24,292,000,OOO 5, 258, OOO.OOO 10,757,000,000 Installation Cost - Land - Study & Supervision - Loan Interest Rupiah Rupiah Rupiah Rupiah 40,307,000,000 1,955,000,000 4,000,000,000 2,400,000,000 40,307,000,000 1,955,000,000 4,000,000,000 3,600,000,000 40,307,000,000 1,955,000,000 4,000,000,000 4,800,000,OOO TOTAL INVESTMENT Rupiah 48^662^000^00 49J962J000^000 51,062^00^000 3. OPERATION COST Capital Cost Fixed Cost Rp/year 4,233,594,000 5,495,820,000 6,842,308,000 - Machines 3Z/yr - Elec. Equip. 1.5Z/yr - Building IZ/yr Rp/year Rp/year Rp/year 728,760,000 78,870,000 107,570,000 728,760,000 78,870,000 107,570,000 728,760,000 78,870,000 107,570,000 Maintenance - Personnel - Administration - Insurance Rp/year Rp/year Rp/year Rp/year 915,200,000 144,000,000 14,400,000 120,621,000 915,200,000 144,000,000 14,400,000 120,621,000 915,200,000 144,000,000 14,400,000 120,621,000 Total Fixed Cost Variable Cost Rp/year 1,194,521,000 1,194,521,000 1,194,521,000 - Fuel - Slag Rp/year Rp/year 68,000,000 120,000,000 68,000,000 120,000,000 68,000,000 120,000,000 Total Variable Cost TOTAL OPERATION COST 188,000,000 188,000,000 188,000,000 Rp/year 5,616,115,000 6,878,341,000 8,224,829,000

- 31 - ANNEX I (continued) Interest Rate 6Z/Year 92/Year 12Z/Year 4. REVENUE Electricity sell price alternatives: Rp. Rp. Rp. Rp. Rp. 24. 28. 32. 36. 42. -/kWh -/kWh -/kWh -/kWh -/kWh Rp/year Rp/year Rp/year Rp/year Rp/year 3,696 2,240 2,560 2,880 3,360 ,115 ,000 ,000 ,000 ,000 ,000 ,000 ,000 ,000 ,000 4 2 ,950,341 ,240,000 ,560,000 ,880,000 ,360,000 ,000 ,000 ,000 ,000 ,000 6,304 2,240 2,560 2,880 3,360 ,820,000 ,000,000 ,000,000 ,000,000 ,000,000 2 2 5. DEFICIT 3 Rp. 24. -/kWh Rp/year Rp/tons 1,920 ,000 9 ,000 ,240 1 ,920,000 12 ,000 ,396 1,920 ,000,000 15,762 Rp. 28. -/kWh Rp/year Rp/tons 3,376 .115 8 ,000 ,440 4 ,638,341 11 ,000 .596 5,984 ,829,000 14,962 Rp. 32. -/kWh Rp/year Rp/tons 3,056 ,115 7 ,000 ,640 4 ,318,341 10 ,000 ,796 5,664 ,829,000 14,162 Rp. 36. -/kWh Rp/year Rp/tons 2,736 ,115 6 ,000 ,840 3 ,998,341 9 ,000 ,996 5,344 ,829,000 13,362 Rp. 42. -/kWh Rp/year Rp/tons 2,256 ,115 5 ,000 ,640 3 ,518,341 8 ,000 ,976 4,864 ,829,000 12,162

- 32 - ANNEX J OPERATIONAL COSTS OF THE WASTE INCINERATION PLANT OVER THE PERIOD OF OPERATION ii D net npcnitjnn costs V cap1t.i! costs Constant A trait oprration costi (fli « ciirrrnt costs » 10 \ ».«) O r»»enuei (electricity) (» 10 \ p..) -»- LIVE TIME HUlt nPEMTICM Tl* or T1 ') ttjtn i The calcul.iiion is based on UK 9t per year loan interest and l^>. 26/KUh.

SYSTEMS ANALYSIS AND SOLID WASTE MANAGEMENT PLANNING Abraham Michaels Consulting Engineer, Osterville, Massachusetts NRC Panelist The purpose of systems analysis is to achieve optimum utilization of resources using a systematic procedure for identifying and analyzing the elements that compose a total system. From this analysis, models of the system are developed, relevant data are collected, and elements of the system are studied in relation to each other and the total system within the control of the management body concerned. Factors outside the control are known as constraints. Systems analysis does not necessarily require advanced mathematics or the use of computers. Where the number of variables is small, arithmetic and human judgment are sufficient. Systems analysis can assist in optimizing vehicle usage, crew size, and routing; locating transfer points; selecting process and disposal methods including site selection; and testing policies and decision criteria before implementation. System constraints that should receive particular attention in Indonesia include: Climate Seasonal variation Financial factors Regional economy and process technology Environmental standards Public health and public awareness Management capability available Technical capacity Land use and availability Physical characteristics of the city including road system Social and religious customs. In addition to these constraints, basic policies may need to be considered such as the necessity for labor-intensive systems to provide employment or existing practices such as scavenging or the use of waste as a soil conditioner. Variables to be considered for analysis should include but not necessarily be limited to the following: • Number, type, and size of container and collection dump sites • Service density - 33 -

- 34 - Topography and climate Frequency and method of collection Crew size Vehicle accessibility condition Vehicle types and sizes Type of waste (residential, food market, commercial, etc.) Changes in living habits and waste quality and quantity Resource recovery return variables. Functions in which systems analysis can offer maximum results are: • Refuse collection—crew size in relation to vehicle type and size • Selection of transport, process, and disposal methods, including sites and size of operation • Determination of types, sizes, and locations of containers • Allocation of vehicle routes. The successful implementation of new systems may require pilot programs to test systems and train personnel as well as incentives— financial, fringe benefit, status improvement—to employees to accept new and more efficient methods. Education of the public may also be necessary. Users and purchasers of recovered energy or recyclables may have to be surveyed and long-term contracts established with users of recovered products.

PUBLIC TRANSPORT IN THE EASTERN CORRIDOR OF JAKARTA Transportation System Study Group,* Directorate for Systems Analyses, Agency for the Assessment and Application of Technology (BPPT) This study of public transport in the Eastern Corridor of Jakarta is but one of the studies conducted by the Transportation System Study Group, under BPPT's Directorate for Systems Analyses. As only a preliminary study, it will serve as the basis for a subsequent study of the entire public transport system, including an improved bus service system in the Eastern Corridor. This study was based on the hypothesis that the level of service of public transport in Jakarta, specifically those that have fixed routes, has deteriorated. In this chapter, public transport is defined as that which has fixed routes, unless indicated otherwise. What follows is a description of the characteristics of public transport in the Eastern Corridor, such as number of passengers, fare charged, daily trips, intervals between arrivals of vehicles, passenger waiting time, and length of trips. BACKGROUND, PURPOSE, AND METHODS OF THE STUDY Background In the last few years, the population of Jakarta has increased rapidly. From 1971 to 1976, it increased at a rate of 4.48 percent each year, 2.5 percent of which was natural and 1.39 percent of which was due to migration. In 1978, the population of Jakarta was 6,081,963 (Special Government of Jakarta, 1979). The number of motor vehicles, especially privately owned, has also increased, at an average rate of 14 percent a year. These increases in population and vehicles have not been matched by the construction of additional roads, resulting in traffic jams in various areas and deteriorating services. Consequently, the cost of operating a vehicle has risen, as well as the demand for public transport services. Public transport plays a significant role in the lives of the people of Jakarta, most of whom have low incomes. In 1978, it was estimated that 5 million trips were made each day: 54.8 percent by public transport, 0.5 percent by train, and 44.7 percent by private car. *BPPT Transportation System Study Group: Ir. Hernowo, Leader; Ir. Arinatjandra; Ir. Indrayatl; Drs. Insanto; and Dr. Soemanti. - 35 -

- 36 - Although public transport is needed by the majority of the people, it has not increased as much as the number of private cars. This is evident during peak morning and afternoon travel periods when city buses always have more passengers than they are allowed, and passengers can be seen scrambling for a place in the bus. Purpose of the Study The purpose of this study of public transport in the Eastern Corridor is to try to solve the main problems it faces from day to day such as: • Inadequate space for passengers using public transport, especially during the morning and afternoon peak hours • Lack of direct routes (without having to change vehicles) from place of origin to destination for passengers using public transport • Need to satisfy the rising demand for public transportation service, especially main road transportation. Methods of Study To obtain the necessary information, data were collected by interview and observation. Interviews were conducted to ascertain the point of origin and destination of passengers along the Eastern Corridor. Observation was used to determine passenger waiting time at the bus stop, number of passengers in public transport vehicles, and the length of intervals between arrivals of public transport vehicles at bus stops (headway). CHARACTERISTICS OF THE AREA OF STUDY Location of Study The study area, the so-called Eastern Corridor, is located in the eastern part of Jakarta, stretching from north to south (see Figure 1). On the north it is bordered by the Lapangan Banteng terminal and on the south by the Cililitan terminal. Main Roads Along this corridor there are nine main roads, some of which are used only for one-way traffic. Thus the traffic from Cililitan to Lapangan Banteng (northbound) and vice versa can choose from a number of different routes. Table 1 lists the names, lengths, and number of lanes of the main roads connecting the Lapangan Banteng and Cililitan terminals. It is estimated that the length of the roads (seven in all) running north and

- 37 - Eastern Corridor Terminal 1.J1 Senen Raya 2.J1 Pasar Senen , 3.J1 -Kramat Raya 4.J1 Sa1emba Raya 5.J1 Hatraman Raya 6.J1 Jatinegara Barat 7.J1 Jatinegara Timur , 8.J1 Otto Iskandardinata 9.J1 Dewi Sartika FIGURE 1 DKI Jakarta area map and the location of the Eastern Corridor.

- 38 - south (including the roads entering the terminal) are 12 km and 12.6 km, respectively. Along this corridor there are 12 ground crossings and 1 two-level crossing. Land Use Land along the Eastern Corridor is generally used for planned housing, unplanned housing, public facility buildings, and businesses (see Figure 2). Most of the Eastern Corridor consists of residential areas, largely in the form of one-story houses. Multistory buildings along the corridor are generally used as public facilities such as government and private offices, schools and universities, and hospitals, and for business or trade such as markets, shopping centers, and shops. Three business centers that stand out are in Senen, Jatlnegara, and Cawang. Alongside the Salemba Raya, Kramat Raya, and part of Senen Raya roads are public facility buildings. Buildings or locations that are well known along the Eastern Corridor and often used as points of reference are: The Finance Ministry at Lapangan Banteng The shopping center at Senen The University of Indonesia complex at Salemba Raya St. Carolus Hospital at Salemba Raya The Ministry of Agriculture at Salemba Raya The shopping center at Jatinegara The East Jakarta Mayor's office at Jatlnegara Timur The East Jakarta Youth Center at Otto Iskandardlnata. Bus Stops and Terminals Along the corridor there are 64 bus stops—30 .northbound and 34 south- bound. The distance between bus stops is not always the same, the closest and the farthest being 120 m and 850 m, respectively. Public transport vehicles must stop at determined points; mlcrobuses and microlets/opelets (12-passenger vehicles that serve exact routes) may stop at any of the allowed points at the request of passengers. There are four official public transport terminals along the Eastern Corridor at Cllilitan, Lapangan Banteng, Kampung Melayu, and Senen. The unofficial terminal is located at Jatlnegara Timur Raya. CHARACTERISTICS OF THE PUBLIC TRANSPORT SERVICE Types of Vehicles and Passenger Capacity Almost all types of public transport vehicles and private cars are allowed to pass through the Eastern Corridor. Public transport vehicles include those with fixed routes such as city buses, microbuses, opelets/microlets, and bemos (3-wheeled motorized vehicles), and those

- 39 - 1 9 O J3 M M I > > M M Ml 3 M > | M MM | & § > M •Q CO - c .0 B u .rl Jg | M M M 1 > > M i M I-l jg M M I M > 1 M .H B VI M > .-4 -H 14 c» X SS S I A «"HI V a « VI 4 4 60 4 4 4 B r-< 4) 4 4) 01 41 4 C 4J B 6 B (Q M 4 B « 44 B i-l O i-l i-l i-l a , .H M O M U B° " ff 2 B° CO a 4-1-1—1 -r< -H -H i-l S. 0. -S + -S -S *S 0\ jS S. » S. p. o. s\ + 4) 4) 41 41 41 i-l 00 CM B C CM C CM (M BB a B 4 « 4 44 • . H <S + t-l r-4 + ,-!+ +. l-ll-l o\ 4J U .OBMC4B MB a MM v40B4 C 2 * S S ON § .o j2 -j< '-a 'j* -o -a j< •j< rH 4 o u S B S S3 g SB B S S VI M O fX O* C^» Of CX 4 Tl * ja a js js M W .ft -H -H 41 -H 0) -H « *-> 4 4 41 X X X X Xr-i X X'H Xr-l X X a 44444 44 4 44 Q 4 r r r r ft r ri I| r r * CM CM i-l i-l CM4 i-l CM OB CM HI i-l i-H 4J 4J 4J .o o 4 4 4 4 4O 4 4O 4O 4 4 4J h 0) 01 41 41 «h ttl 4) W «)M 4) 41 M • i B B B B BUH B B<M B«H B B a -i ,-! ,-| ^H i^ i-l(M i-l i-lCM i-ICM; i-l rH 3 ^ * CM %o .* *\ <«+ m *.». \o+ m ir> H IM rH 4) O h 4 M m CM in o co u-i r-i m o o «oo'irtOi-l <t o m oo H « -3 CMrH<tCMCO IA-* <M i-IO •1 JS 3 CMCM»-li-li-l i-l 1-4 1-4r-I 1-3 .o 4E a M 4 Q 43" 4 4 „ 4J 4 X X 4 &• 4J 4 4 4J 0) 4 a s J^ ,J .H U M 4 0 2 g 4) M 4 .H 4 4 « M H X a M SB .H B 4 4 (S « X 4« 4J4kiM 05 0) X a MJ444B ri4W 44MM4 4 B! CO CO M 4) « B -a VI B B 4 B 9 B h a 3 .H O -H -H b « B v 4 i 5 54J4JU4J i-l 4 Cm 4J444 « h «4 § H OO^^lSB co M copLi M

- 40 - Unplanned Residential area Planned Residential area Public building Commercial area FIGURE 2 Existing land use along the Eastern Corridor.

- 41 - without fixed routes such as taxis, public transport vehicles type IV, and becaks (S^wheeled, human-powered vehicles). Minibuses and pick-ups are unofficial public transport vehicles and have fixed or unfixed routes according to the requests of passengers. Both large and medium city buses operate in the corridor. Each public transport vehicle has a predetermined route which is signified by an exhibited code number. Passenger capacity is defined as the number of passengers allowed according to the space available. All official public transport vehicles carry signs showing their passenger capacity. Table 2 shows the passenger capacity and number of seats available on each kind of public transport vehicle operating in the Eastern Corridor. Fare Some vehicles charge the same fare for a single trip regardless of the distance traveled, while others charge according to the actual distance traveled. Each means of transportation charges one fare for one trip which means that the more frequently a passenger changes vehicles, the more he pays. Table 3 shows the fare charged each passenger for one trip. Routes and Characteristics of Passenger Trips The routes of city buses and opelets/microlets are shown in Figure 3. In the morning, the number of passengers heading north is generally larger than the number heading south, while in the afternoon the opposite is true. This is explained by the fact that most activities are centered in the north, while the south is composed mostly of residential areas. Morning peak hours are defined here as 6:00-9:00 a.m.; morning off-peak hours are 9:00 a.m. to 12:00 p.m. Daily Trips The Jakarta Metropolitan Area Transportation Study (JMATS) has estimated that from 1972 to 1974, 3.6 million person-trips were taken daily in Jakarta using public transport and private vehicles (60.8 percent and 39.2 percent, respectively). Of this number, 619,000 trips were made in the Eastern Corridor. Figures compiled by the BPPT show that in 1978 the number of daily trips taken in Jakarta reached 4.995 million— 55 percent by public transport and 45 percent by private car. These changes have resulted from the increase in population and a rise in the number of private cars in Jakarta. Using data from previous studies, it has been established that daily trips taken in the Eastern Corridor in 1978 rose to 20 percent of all dally trips taken in Jakarta. Based on the 1978 figures, it can be estimated that the number of daily trips taken in the Eastern Corridor in 1980 was 667,800 using public transport and 546,380 using private cars.

-42- TABLE 2 Available Seats and Passenger Capacity of Public Transport Vehicles Seats Passenger Type of Route Type of Vehicle Available Capacity Fixed City bus (large) 45 65 City bus (medium) 32 50 Microbus 20 20 Microlet 11 11 Opelet 7 7 Bemo 7 7 Free Taxi 4 4 Three-wheeled 2 2 Becak 2 2 Number of Passengers By observing the average number of passengers on city bus route no. 40 and opelets/microlets, it was possible to determine the average space available in these vehicles (see Table 4). Figures 4 and 5 show the fluctuation in the space available on city buses and opelets/microlets during peak and off-peak hours in the morning northbound; the same is shown for southbound in Figures 6 and 7. An estimate of the number of passengers on all public transport going through the Eastern Corridor was made on the basis of direct observation sampling. Total number of passengers in vehicles on different roads, size of vehicles, flow of traffic, and times of vehicle operation were recorded. Figures 8 and 9, which show the density of passengers on public transport vehicles on every main road, were com- piled according to the direction traveled and divided in half, for peak and off-peak hours. Arrival of Vehicles on Each Road and Headway Table 5 shows the number of arrivals of city buses, microbuses, and opelets/microlets on each road. Interestingly, the number of city bus arrivals is proportionately the opposite of the number of arrivals of opelets/microlets. Thus apparently the opelet/microlet fleet fills in the gap in city buses. Public transport does not arrive at regular intervals at bus stops along the Eastern Corridor. It has been found that intervals on the northbound route are longer than those on the southbound route.

- 43 - TABLE 3 Type of Public Transport and Fare Charged per Trip Type of Type of Route Vehicle Minimum Maximum Fare (Rp) Fare (Rp) Explanation Fixed City bus 50 Microbus 50 Opelet/Microlet 50 Bemo 50 50 50 125 75 Per one trip close/distant. For students the charge per trip is Rp. 30. Same as above Fare depends on distance traveled. Same as above Free Taxi 250 Three- wheeled 150 Becak 100 Rp. 250 for the first km and Rp. 120 for every additional km. Total fare is determined by meter. Fare charged depends on dis- tance traveled and bargaining between passenger and driver. Same as above TABLE 4 Average Space Available on a City Bus (Route No. 40) and Opelet/Microlet Direction City Bus Opelet/Microlet Peak Off-peak Peak Off-peak North South -19 0 +2 4« -1 -5 -1 -3

- 44 - i i! 530 fl m i FIGURE 3 Number of public transport routes in the Eastern Corridor.

- 45 - IM Li/.J_L. 1 /I i i I !__ ; !. . i . : -' 1 ' '. I ' ; !' 5 ^^ \ ' . ! - ; '_! /' : I . ! ! / ; ='" t i 'I i i _ t _L_:_i i « ft j_. ' ; j-^ :.:/.. ~r~~: .'; - —""/ •'. ' -:—r—:— t i 111 J -10^-^- I I I ! i i . I J_L->v < > * u <. a /t ^I ! ' i i i i 1 t i . i i i I I -' » i i i i i i i ! I i i i i L t SPACE- AVAIL ABLE IN- PUBLIC-TRANSPORT" VEHICLES DURING- MORNING- PEAK HOURS" r— ; — ?- i. NOTATION j -----H-H-r- I — . . —.. . „ .. . -.•- .mfi m , : B U S ! : L ' ' ' i i 1T: . . i I . : . 'ii'l •1OPELET-.; . -. : ''? "L APANGAN ~ BANTENG'- TEHMINAL" '« OLILITAN'-TERMINAL' -:' K A MPUNG" ' ME L AYU TERMINAL " i '• • « n 'a n ' s « n » » 20 21 72 a BUS STOP NUMBER FIGURE 4 Space available in public transport vehicles during morning peak hours, northbound.

.10 m < OL in ~n ' 1 1 ! : i i ! • 1 I : i | l '. | ! . j 1 'If! 1 '. : i : i : i i S. r~j '.... i i i ! I/ ! I : . ? ' ! i i : ! ! ; , i ; i /i 1 : | i ' i ! \ i : • i i .: . I 1 . i . \l . ' i 1 i 1 1 | ' i , ! 1 ' 1 ' 1 I : ! * 1 I | V i : 1 ! ! ' - ! '/. , i : \ . i i : .< » . ;..;....., /: . i \ i i ; : ; 1 ' i i . : : . . / 1 \ . ' i . ' ... 3 - = . . / - i . ! | V . : . . . • ! : 1 : / . ! \' 1 . : ! | : • : / \ • / ! \ -\— r '.'" ' j _... \. .;.!.. . .-. • : / \ . : : ! . S \ 1 .:.,." 1 . . ! ' X . ! ' ^ . . . ^ ; :Si S^^L.-^, _ — -~~*~ , j , I i 1 . • t : . . : : ! ! I . | ; i • . Pj i i ! ; ! ; . : • ; ! iii : : . . i . i j i i , ; | • ! ! .1 . • . i | \ ! , * ' ' ! ! : i . : . i i ; i ! 1 J _ j l i'.-'., i i : i . : . : 1 ! i ! ' ! i 1 i ! . . ! . ! 1 ' ; f i! : ' . -. 1 ! ' ' . : i . ! • ' : : 1 i 1 . j i I : -I , ; i ! : ' i : ' . . | . , ! 1 i ; i i i i i ' ' J - : i i i i i . ' I- J . . I I '1 .1 T.T I i I -- SPACE -AWIL/BLE IN PUBLIC—TR ANSPORT- VEHICLES - ~^MQRNiN6~QFF"PEAK~HQURS~ -30— DIRECTION" «i" NORTHBOUND : '~ '. . 1 : . : : . ' i NOTATION OPE L E:T » LAPANGAN" BANTENG - •-C!LiLITAN'_TERMIN,AL"rr - '. ' T KAMPUNG MELAYU" "TERMINAL -30- (. ' v aa c M i? » » x a a * TS » 77 BUS STOP NUMBER FIGURE 5 Space available in public transport vehicles during morning off-peak hours, northbound.

: . ! M. ! i ! i . i i n s..: >*» i > t ^*r ±0- . i I .1 I . ill! I I i i i i ! i ! ^4J- 4-U.-J-: • : i MM i i i I ; : III;;.. I I 4.1. '\~r ' I I r i —SFfcCE-AVAIbABLE- IN; PUBLIC-TRANSPORTf VEHICLES -IOHDIRECTION~~SOUTHBOUND: NOTATION . !. - j_ __; _ -rjr----j..B.u s. -...:_L ' LJ_Q P E LET. «KAMPUN(5 MELAYU TEnMiNAL -20- vnn BUS STOP NUMBER FIGURE 6 Space available in public transport vehicles during morning peak hours, southbound.

- 48 - u u :3o- ; . :_it. i i , i I LI ill'1! I p i : ' I i ! -; m-r-rr '. ' i * i I J Li ' ' * .Li ' ! . ! I ' I i ' ' ' • ~^^iDj±tb:Li:rt :m r .' i. ITT 'ji.ii.jr . : . '. : . . ' ' i •; ' ;J . ! Li Li.-i'IT - ' ' . i ' . i...!;: : j ! : :.!'•.:I ! ! i ! . f fTT i I I | ;" ri .~• I i i i i~» r» . r:3 -U-! ' ppJ-J-jlji i TTri-LLJ..i..Lr?l;.i.8. l..i :.4_"!x> . ; Hi^p; T^ . i ; t j . i j ; - ' . . 1 :.|.. i.; , /. n iii i . ; . -r"~~~T^i'TT'r:" r"'rr 'i ! '/. "" «2o'^---—.:-T-i-;-:---;--'------f7r'-rV' r T "\. j ' 'I . i"1 '. : ". '~r'-' r . ; « .•;•-. y .. , ~ - -—'-—-~ -A—— i .. i ., i . , i-..-i-.i.-^-^.-. ..-. .. iii { i I Jill I i i : II | rrrn'r I ! I U-J n- —SPACE- AVAILABLEHN- PUBLIC-TRANSPORT" VEHICLES HLDURINGZLMORNINGT: OFF.;.PEAK' . ".H.OURSITTZT '^ZZDiRECTIONHZTSOaTHBOUNDltZ: NOTATION ; { ^.i . ^ ! LjLJLJ.l-l£u ~i i__ "_i"jL, | ]" ":""" "~ • fl B U S1 .- . —. .-... _ J OPELET .' "T4 -~"LAPANGA'N' BANTENG TERMINAL T'CILILITAN 'TERMINAL'1 "O " ' «" KAMPUNG ' MELAYU-" TERMINAL'' -r I 1 3 4 | • ff ft V • • • • • • V • • • • • • U 1^ BUS STOP NUMBER - FIGURE 7 Space available in public transport vehicles during morning off-peak hours, southbound.

- 49 - «• o o C 0) 4 CO CO QJ C 10 S H § vo m i-4 01 in * i-H O • vo ^ .V in *> in lo VD I(N O •> ""1 O o» in Irn o «•» R o> m MI CM CO o CM in rt .« M 3 CD m at H to 2 vo H H n i CM o o\ r» n 3 in vo H H 3 CM 3 1 i vo TJ- M tl O 0 CO er\ -H pJ H ( 1 >l H "W r- r^ iH - ft c cc vo ^r <N O CC i VD .» CN CM CM CN 04 o a. ) VO ,* ' CN C 3° 3 r m f> n in CM *( Ff nr : i 1 . ta asuan a abuas q *A * 1 u & O o K 4J .H O 8> 8, 8 £ * 2 D o\ in ,r rH vo CO 0 .1-l M rH P CM 9 H in" en • r- co CO <* ^ « ^ m CM in VO an n vo . rH ^ vo I O i^ . H in r <l CM > o in P .4 ^^ B CM r H rH ri -r-4 o\ vo n Q E^ in 2 ^ £ ! CM rH f-f *^ - in ^ * 3 I- rH O o vo 00 CM H n m H ^* CN * of CM in rH cn H in cT CM VO &> en c TT O, C ft) H CTN m 4-J i— | af FIGURE 8 Passenger density of buses and opelets/microlets traveling on seven main roads during morning peak hours.

- 50 - M B £ iJ s S S s s * s 0. B H "rt CN O\ ^ j pa o £. f? *°, 15 32 o « 7 °° ^ 2! "'if s c P? iH > § v *J o OB ^J, Q. -H « • o u vo ,0 o i i FTP Q c o 5 in ^ ^ tt SI , rH ^ H 0 23 00 vo tN z (J ft. oo r~i ""J. M M H en' £} ^ D,M H H tc » — ' K Q vo S " j. H H S S 51 "*. °. -"- i-< H rj lO VO .H o oovo^rrMooovo ^'rMooovo ^MC O t I « i 3 r- 3 M S. R ° . H H 00 .H U ^ -s? ^r s 1 § j! s a' a » r» r» vo r H ID A > S *j u £ r? ° ^ i ro <N ess * O ro j= s i a a 1 vo ^ o H vo vo H H (N n o p I DC H fN ??s. U FIGURE 9 Passenger density of buses and opelets/microlets traveling on seven main roads during morning off-peak hours.

- 51 - TABLE 5 Number of Arrivals of City Buses, Microbuses, and Opelets/Mlcrolets City Bus Microbus Opelet/Microlet Main Road, Northbound Off- peak Peak Off- Off- peak Peak peak Peak I 140 166 — — — ~ II 191 182 — — — III 352 305 — 582 656 I? 267 238 — — 584 667 V 401 339 403 384 1,254 1,410 VI 422 339 — — 727 746 VII . ~ ""• 3 9 585 593 Main Road, Southbound VII 205 218 — — — — VI 232 200 — — 756 641 V 283 256 — 605 626 IV 243 263 — — 636 665 III 420 435 452 433 1,282 1,418 II 415 422 — — 611 612 I -- •w 3 9 534 558 Passenger Waiting Time Length of waiting time is the time needed for a prospective passenger to get on the public transport vehicle of his choice, counted from the time the prospective passenger arrives at the bus stop. The primary factor influencing the length of wait is the length of the interval before arrival of a vehicle and the space available in the vehicle. Other factors include the patience of the prospective

- 52 - passenger, his readiness to crowd in the public transport vehicle, and the possibility of his choosing another type of public transport vehicle. Based on observation, it was determined that for prospective passengers in the Eastern Corridor the difference in the length of wait for city buses from the major fleet compared to the minor fleet was insignificant. It was also learned that during peak hours, the length of wait for prospective northbound passengers was shorter than that for those headed south. The opposite is true during off-peak hours. The length of the intervals between arrival of vehicles has a greater effect on the waiting period than the space available in vehicles which has very little effect. Length of Trips The length of trips for vehicles in the Eastern Corridor is the time needed for the vehicle to: • Travel from the terminal of departure to the terminal of arrival • Or to enter the Eastern Corridor and arrive at the arrival terminal in the Eastern Corridor • Or to travel from the terminal of departure until leaving the Eastern Corridor area. Table 6 shows the results of data gathered on the length of trips and speed of public transport as well as the flow of traffic in the Eastern Corridor. The length of trips made by public transport vehicles is estimated to be 1.6 times the flow of traffic as a result of the frequent stops made by public transport at predetermined points to pick up or drop off passengers. CONCLUSIONS Based on studies, observations, and interviews related to public transport—city buses, opelets/microlets, and microbuses—it can be concluded for the case of the Eastern Corridor that the rise in demand for public transport service is not being met by the number of public transport vehicles providing the service. This is especially notice- able in the demand for city bus service during peak hours. In certain situations, public transport has to carry more passengers than are allowed. This condition requires the appropriate attention, and it should not be allowed to continue without improvement in the public transport system. In weighing possibilities for improving the system, their positive and negative effects on the system must be considered. Some possible structural changes include:

- 53 - TABLE 6 Travel Time, Average Speed, and Length of Trips for City Buses, Opelets/Microlets, and Traffic Flow Northbound Travel Time Speed (min.-sec. ) (km/h) Kind of Off- Off- Distance Vehicle Route Peak peak Peak peak (km) City bus Cililitan- 38' 5" 35 '45" 19 21 11.96 Lapangan Banteng Pasar Minggu- 12 '44" 11' 5" 20 23 4.20 Opelet/ Kampung Melayu microlet Kampung Melayu- 24' 2" 24 '12" 18 17 7.00 Senen Traffic Cililitan- 26' 17" 21 '24" 27 33 11.96 flow Lapangan Banteng Southbound City bus Lapangan Banteng 35 '15" 37 '50" 21 19 12.05 -Cililitan Senen- 17' 5" 19 '42" 18 16 5.10 Opelet/ Kampung Melayu microlet Kampung Melayu- 11 '55" 12' 20 21 4.20 Pasar Minggu Traffic Lapangan Banteng- 25 '24" 22 '26" 28 32 12.05 flow Cililitan SOURCES : BPP Teknologi Observed; DKI Jakarta, DLLAJR, Travel Time Study Kota Jakarta 1979. 1979/1980.

- 54 - • Adding to the number of city buses in the armada • Adding to the space available within a city bus unit (with double-decker buses) • The use of a rapid and limited bus system. One infrastruetural change might be widening the roads in a few areas in the Eastern Corridor. REFERENCES Special Government of Jakarta. 1979. Third Five Year Development Plan, 1979/1980-1983/1984, p. 225. Special Government of Jakarta. 1979. Transportation Department, Travel Time Study for the Metropolitan City of Jakarta, 1979, 1979/80. BPP Teknologi Observed.

SYSTEMS ANALYSIS AND A STUDY OF URBAN TRANSPORTATION PROBLEMS Britton Harris Professor, School of Public and Urban Policy, University of Pennsylvania NRC Panelist The rate of urbanization is very rapid in many developing countries, including Indonesia. The City of Jakarta now has a total population of about 6 million persons, and other cities, especially in regions of overall high population density such as Java, are continuing to grow rapidly. The economic health of urban concentrations depends on the division of labor and the effective organization of production and consumption. This organization requires the interaction of workers and households as producers and consumers with economic establishments that manufacture goods and produce services, thereby assisting the activities of other companies and households. The assembly of a labor force and of customers and the distribution of goods and services within the metropolitan region and beyond its borders imply various means of interaction. Interaction is provided largely by means of transportation, although on occasion electronic and written communication can substitute for transport. There is no substitute for transportation, however, where the assembly of people and the distribution of goods are concerned. This chapter focuses largely on the transportation of persons within the metropolitan region, although the transport of goods should not by any means be neglected and will be occasionally touched on briefly. From an economic point of view, the transport of people related to pro- duction and consumption processes is economically far more costly than most goods transport, primarily because of the perishability of the commodity being conserved—human time and effort—as well as considera- tions of comfort and safety. At the same time, the condition of the transport system in relation to all other aspects of production and consumption has a very powerful influence on the well-being and comfort of the residents of the metropolitan region. When conflicts between goods movements and persons movement arise in the metropolitan region, it is frequently possible to redistribute goods movement in time and thereby relieve the congestion that it generates and that adversely affects the movement of people. In ancient Rome, roadways became so congested that goods movements were restricted to a few hours in the middle of the night. In this chapter, the nature of a systems analysis of metropolitan transport will be taken up from several different points of view. - 55 -

- 56 - First, the nature of the system will be noted, followed by an examin- ation of the behavioral aspects of the uses of the system. Finally, there will be a brief discussion of the technological means that are available in computer software to analyze the system's problems as well as a general overview of the process of systems analysis of transport needs. THE NATURE OF TRANSPORT SYSTEMS A transportation system consists of three major components: (1) the technical or machine system components consisting of vehicles, guide- ways, terminals, interchanges, etc.; (2) the users of the transport system, together with their needs, desires, and modes of behavior; and (3) an institutional system that provides for the management of systems and the channeling of behavior. According to its varying degrees of articulation, this system may heighten the efficiency of the system and minimize the adverse effects to be expected from congestion and conflict of uses. The different types of equipment and channels of movement are generally called modes of transportation; the principal modes may be distinguished as water-borne, air-borne, and land transportation. These could be subdivided into subsidiary types of movement; for example, water-borne movement can be by sea or by inland waterway, and ground transport can be on foot, in a variety of different vehicles that travel on roadways, by different types of rail transport, etc. The institu- tional setting provides that several of these different modes can be publicly or privately provided, or both. The behavior of different modes will depend jointly on their technical characteristics, their user behavior, and the institutional management. Transport activities generally connect separated activities by way of a continuous route. There is, of course, an exception for air trans- port, where routes are usually continuous everywhere unless extremely high mountains or completely adverse weather intervenes. Typically, long continuous routes are made up of segments that differ in their technical operating characteristics and in their use. Also typically, in the urban setting, routes that connect points in different directions intersect each other and consequently mutually interfere with the free flow of movement. Because of the evolution of the use of transport systems, and a concomitant evolution of the systems themselves and of transportation analysis methods, it is commonplace to refer to large classes of trans- port movement as movements in corridors. The most elementary system in transportation analysis, if not a single link, is at most a corridor of movement. Here one encounters all of the behavioral problems of individuals utilizing and participating in the flow of movement, the conflicts of different modes of travel, and the potential conflict between corridors at points of intersection. Finally, solutions to problems for corridor movement can usually be formulated in terms of relatively simple alternatives that can readily be subjected to a cost-benefit analysis.

- 57 - i' At a higher level of complexity, the transportation system must be regarded as a network. At this point, the problems of representation, analysis, and problem solving become more difficult by an order of magnitude. Within a network, huge numbers of alternative paths can be taken by unconstrained vehicles such as private automobiles, taxis, and bicycles, and there are many points of extreme congestion owing to the crossing of routes. In addition, the number of potential inter- connections between different pairs of points in a metropolitan region is greatly increased, largely because the number of points being considered is also increased: the interconnections rise with the square of the number of points. From the point of view of system performance it must always be remembered that transportation is an intermediate good that is designed to produce other final goods such as efficiency in production and effectiveness or enjoyable aspects in consumption. For this reason, one must consider the social and economic system that the transport system is designed to serve. This is particularly true because the behaviors that can be identified in response to the state of a transportation system involve more than the day-to-day behavior in choice of mode, route, and destination. Adaptations over the longer run can take the form of the relocation of places of residence and places of economic activity. In the larger sense the transportation system contains a network of facilities of many different modes, some of which include the same rights of way and some of which do not, imbedded in a system of economic and social interaction in which many individual decision makers take actions. These actions in turn influence the state of the trans- portation system and this state influences future actions. MODALITIES OF BEHAVIOR IN THE TRANSPORTATION SYSTEM Different kinds of human behavior on the part of individuals, house- holds, and economic establishments result in the differential perfor- mance of transportation systems. The framework for discussion of these behaviors comes from the standard terminology and modeling procedures used in transportation system analysis in the United States. This must be done with some caution, however, because the types of behavior that will be encountered in Jakarta or Surabaya are not altogether properly captured with U.S. analytic practice. Yet adaptations in this practice can be made and will be discussed briefly. In a static situation where the status of the transportation system is known, the average user of the system in making a decision does not conclude that his behavior will change the state of the system. An exception to this, however, could be noted in the event that a large entrepreneur or government office was considering a new establishment on currently vacant land and had to take into account the impact that the load it will generate would place on the transportation system. In general, it is assumed that individual decision makers behave as if their activities would not affect the system. Of course over time, and given a large number of such decisions, this expectation frequently will not hold.

- 58 - The necessary external connections that a household, factory, or office must make with other parts of the metropolitan region establish a broad pattern of trip generation. This variable measures the number of trips likely produced by any particular land use, and it depends on the activity levels of the land uses themselves. It may also vary according to the state of the transportation system, so that in a situation where there is a great deal of congestion, many users will find ways to curtail the number of trips that they make and the load that they place on the system. Some trips are less flexible in their adaptation to congestion in this regard than others; for example, work trips are practically obligatory for both the worker and his employer. It should also be noted that trip generation is strongly influenced by cultural conditions; for example, in many parts of the world it is customary for shops and offices to close at midday so that employees can go home for lunch. This doubles the number of work trips in a particular metropolitan area and creates a second set of peak-hour travel demands. Opening and closing times are set by custom, but can be staggered to reduce traffic peaks. The choice of whether or not to make a trip is influenced by other considerations and other choices that have to be made. Perhaps the primary choice after a tentative decision to make a trip is mode of travel. For most trips, there is some choice among walking, cycling, bus, jitney, or private automobile. Such choices are influenced by the length and frequency of a particular kind of trip and the cost of the different kinds of modes—both in time and in money. In addition, some modes are not available in some locations because of the configuration of the transport system (for example, buses) or because of the socio- economic condition of the traveler. Not everyone, for example, can own an automobile or even in some cases a bicycle. Decisions regarding vehicle ownership are made over the long run, often in conjunction with decisions on where to live and where to work. Choices are also made with respect to the destinations of trips. The decision about where to shop, for example, will depend on the distance and attractiveness of various shopping facilities. With an increase in family income and mobility, households will frequently make a smaller number of trips using an automobile, a jitney, or a borrowed car to larger shopping centers that provide lower prices and greater selection of commodities. Decisions about where to go for entertain- ment, visiting friends, and so on are subject to similar influences. The journey to work is obviously more constrained, because the decision to change jobs or to change residential location is not made very fre- quently. Nevertheless, clearly such adjustments are made, because short trips to work are far more prevalent than long ones. In the long run, all of these patterns are influenced by changes in locational patterns. In a growing city like Jakarta, new residential areas are opened and new sites for the establishment of economic activities are likewise developed. These changes result in constant changes in trip generation, choice of mode, choice of destination, and other aspects of travel behavior. The state of the transport system and of the communication system influences this pattern of location and relocation. In the absence of good transport and communication, there

- 59 - is a strong tendency for economic activities to stay close together. This lmposes a considerable extension of the journey to work as the city grows, and leads to additional costs of congestion and lost time in doing business. In this sense, congestion feeds on itself by making decentralization impractical. At the same time, the absence of adequate alternative means of communication becomes very important, for conges- tion could sometimes be overcome by decentralization that is made possible by improved telephone service and other aspects of economic development. Frequently, the types of behavior sketched above are not well understood and some socloeconomic study of them becomes necessary. This kind of study is an underlying activity of the metropolitan transporta- tion studies that have been conducted in the United States, but it is not clear that these extensive survey methods are necessary for the types of problems faced in a typical Indonesian city, or indeed for some of the problems faced in American cities. This is examined further in the concluding section. MATHEMATICAL REPRESENTATIONS OF SYSTEM BEHAVIOR From about 1950 to 1970, the mathematical representation of system behavior and of behavior in the use of transport as sketched in the preceding section developed very rapidly in the United States. Con- sequently, transportation system analysis is probably the best documented and most easily managed of a wide range of computer systems representations. Relevant computer systems are available from a number of consultants and through various U.S. transportation agencies includ- ing the Federal Highway Administration and the Urban Mass Transit Administration. These materials are not specified in detail at this point but are simply sketched. A number of transportation phenomena are susceptible to mathematical and analytical representation in typical operations research style, but these are not discussed in detail in this chapter because generally they do not partake of systems characteristics. They do, however, sometimes lay a basis for small-scale improvements. For example, the stochastic arrival of vehicles at intersections and the behavior of different signalizing systems are well understood. When these signals are extended beyond a particular intersection to a route or corridor, some systems characteristics begin to emerge. Similarly, individual driver behavior in a "car following model" can be analyzed using numerical simulation, differential equations, and other methods. This analysis leads to the conclusion that above certain levels of traffic density, instabilities appear and there are apt to be standing waves of conges- tion even though the highest possible levels of congestion have not yet been reached. Once again this is an isolated phenomenon and not a general system phenomenon. The relationships among the engineering characteristics of guideways, the mix of traffic, the volume of traffic, and the speeds on the links are complex ones that are well understood at an empirical level. These relationships are sometimes used to examine in detail the behavior of transport systems considered as

- 60 - networks. All of these small-scale phenomena are only touched on at this point and will not be the principal objective of the discussion. The most important focus in transportation analysis is on the utilization of transport systems by household decision makers. The three processes of trip generation, choice of mode, and choice of destination (also called trip distribution) are modeled in all of the standard analysis packages. The results of these behaviors are applied to the transportation system through processes known as "tree tracing" and "assignment." Tree tracing finds the shortest paths between all pairs of points on the network, and assignment sees that all movements between pairs of points are attributed to the shortest route or to some combination of best and second best routes between pairs of points. Behavioral models depend in part on the costs of interaction between pairs of points and the availability of services. The two processes of systems analysis and behavioral analysis interact since it is necessary to find the shortest paths in order to determine how households behave, but the behavior of households may lead to congestion on the network and thus change the supposed times. Finding an equilibrium between the behavior and the state of the system is an additional problem or level of systems analysis that can be solved at some added expense. The tracing of the shortest paths in the network and the assignment of trips to different modes and links in the network are greatly simplified if the network consists principally of a corridor with a very small number of parallel transportation routes. In this case, the tracing of routes is extremely simple and, instead of a full-fledged assignment process, one ordinarily engages in a diversion analysis that measures the extent to which trips will be divided between the few existing alternatives. This methodology was well developed even before 1950 and was indeed the principal methodology of transportation analysis before methods for tree tracing and assignment were discovered and applied in large-scale computers about 1955. The analysis of trip generation by different types of household and different types of economic land use is ordinarily based on a survey and regression analysis that takes into account various characteristics of households such as size and income, number of wage earners, and avail- ability of owned vehicles. The analysis for economic establishments involves such matters as the number of employees per square foot of floor space, the character of the activity, and the necessity for direct contact with customers or with the suppliers of services. The analysis of choice of mode and choice of destination is more complex. The general economic model used here is a choice model called by various names depending on its origin. In transportation analysis the model is called a "production-constrained gravity model," while in economics the same model is frequently referred to as a "multinomial logit model." In essence, decision makers are perceived as making choices that are proportional to some function of the attractiveness of different opportunities. This attractiveness is affected by the size and quality of the opportunity as well as by the inconvenience that would be encountered in extended travel. In an established model these calculations are quite straightforward, but choosing a model and estimating its parameters once again require survey materials and, in this case, more effective statistical techniques.

- 61 - A wide variety of statistical survey techniques exists for collect- ing data and deriving the parameters of behavior that are necessary for the operation of all of these models. The simplest and most traditional is a household survey on a sample basis. The old U.S. Bureau of Public Roads standards for these sample surveys involved a very heavy sampling rate, but it is currently believed that (given other information to be discussed below) much smaller samples of perhaps as few as 3,000-5,000 households can provide satisfactory information. Alternatively, surveys can be made at places of business and offices, covering both the employees and customers of these establishments. This form of survey is somewhat more difficult to relate to household behavior, which is critical in the analysis of transport problems. A much more limited and more ad hoc kind of survey material is based on cordon counts and interviews. In this process a number of cordon lines and screen lines are set up in the metropolitan region, and counts are made of vehicles and passengers crossing them, together with roadside interviews of sam- pled individuals. In heavily congested situations the sampling and interviewing are very difficult, and in any case the amount of infor- mation that can be elicited in such an interview is always much smaller than what can be discovered in a household or establishment survey. Nevertheless, in the absence of other feasible methods of survey, adequate information can sometimes be obtained from this source. An entirely different type of information is required in addition to this survey material to operate models once they have been fitted to the data, or calibrated. This information is on the present and future location of activities of various kinds of households in various cate- gories, and information as to levels of income and vehicle ownership that will prevail in different localities during the dates for which the analysis is being made. This kind of information is universal, covering all locators and all households, and it must be exhaustive even on a sketch plan basis if estimates of the response of the total transporta- tion system are to be made. The preparation of these estimates can be undertaken in a very sketchy way by projecting present trends and making guesses as to future locational patterns, or it can be undertaken more systematically through the estimation of the actual locational prefer- ences of households and businesses in future periods of time. These locational preferences would have to be analyzed on the basis of different kinds of surveys from those sketched above, and the operation of models based on these surveys depends on the differential behavior of different transportation plans. This level of analysis is within the reach of most large urban planning agencies within a few years' time, but it is not necessarily congenial with the mode of planning and pros- pective development that is currently in effect. Its adoption is a matter of policy in itself. It must be cautioned that an analysis of transport behavior—if it is to respond to a number of different policies in the analysis phase— must reflect the effects those policies may have on individual behavior. Thus, for example, gasoline taxes, parking taxes, changes in bus fares, and a number of other measures all reflect the economic desirability of shorter or longer trips and of choices between different modes. These choices are not well reflected in a policy analysis if the behavior of

- 62 - households and businesses in response to economic pressures is not well understood, and these matters cannot be well understood if data regard- ing them are not collected in a household survey or other analysis. AN OVERVIEW OF TRANSPORT SYSTEM ANALYSIS What has been sketched thus far lays the basis for a systematic evalua- tion of transportation improvement proposals. The proposals made require careful exploration of the options open to government. These may include the provision of facilities such as roadways, tram lines, and bus lines as well as economic incentives such as taxes and subsidies that have an impact on households and establishments. These policies may also take the form of regulations and enactments that change the supporting management structure of the transportation system by affect- ing the prevalence and operating conditions of buses and jitneys or other forms of public and private transportation. The process of plan- ning transportation improvements will involve setting objectives, out- lining alternative ways of reaching these objectives, and evaluating these alternatives. The experience gained in the planning process will, of course, mean that the objectives and the alternatives will be repeatedly modified in search of more accurate and better embodiments of public policy. The general setting of objectives must be in terms of the desired social and economic outcomes as well as the objectives that can be properly served by a transportation system. These include accessi- bility to employment, a reduction or minimization of interaction costs for both businesses and households, and the effective operation of business establishments. At the same time, public policy probably requires a minimization of investment and operating costs. All of these objectives have to be specified in terms of the operating characteristics of transport systems and of the burdens or benefits that they confer on the users of this system. In specifying alternatives, a serious problem will be encountered with the proliferation of alternatives if several different steps are considered jointly. This can only be avoided by artful planning, but care must be taken not to exclude useful alternatives in the process of imposing limits. The technical methods discussed above consist of computer methods for exploring the consequences of various decisions. The outcomes of those computer models display congestion in various parts of the trans- portation system and the costs to individuals and establishments of making use of the system. If the decisions would result in a curtail- ment of trip making, this too can be taken into account as an incon- venience imposed on users of the system. Similarly, trips that are lengthy either in miles or in time are readily analyzed by these various devices. In short, detailed analyses can be made, and, if the objec- tives are clearly defined, a post-processing phase can be used to sum- marize the impact of each alternative plan on the users of the transport system. The choice between plans then becomes a matter of executive action guided by the inputs of transportation planners and possibly of affected users.

TOWARD A CONCEPTUAL FOOD SYSTEM FLOW MODEL FOR RICE IN INDONESIA Bambang Setiadi Leader, Indonesian Food System Study,* Directorate for Systems Analyses, Agency for the Assessment and Application of Technology (BPPT) The objective of the Indonesian food system study is to build an Indo- nesian food system model for use in assessing the impact of government programs and new technologies on the food system (BPPT and SRI Inter- national 1979). The flow model that has been designed for rice will help Indonesian policymakers evaluate the impact of a wide variety of program alternatives by simulating the impact of food policies and new technologies and determining the optimal policy. Because of the comprehensive nature of the model, it must be applied by many agencies in Indonesia, and this feature of the food system model will provide a base for intra-agency cooperation in food policy develop- ment. Other benefits derived from application of the food system model include: • Training of both systems analysts and food experts through direct participation in policy analysis • Identification of specific problems that need immediate action and will be assigned to short-term task forces • Organization of information and data on agriculture into a manageable format for decision analysis. APPROACH Construction of the flow model for rice began with a general description of the rice system from producer to consumer. This was compiled on the basis of discussions with experts from the universities and research centers, policymakers from Jakarta and region, and "key persons" from regional institutions. To formulate subsystems of the model, it was necessary to enumerate (1) the functions performed by individuals or groups in the flow of rice from producer to consumer, (2) the kinds of *The members of the Indonesian food system study are Anton Gunarto, Daru Mulyono, Djoko Pitojo, Hariadi Wardi, Kasiran, Muhadi, Nenie Yustiningsih, Sri Rustiati, Subiyanto, Sudarmodjo, Sudaryanto, Supriyanto, Syakur Salim, Syalirun Hutagalung, Tri Djoko Wahyoni, Yusuf A. Hidayat. - 63 -

- 64 - commodities handled by them, and (3) the kinds of activities they perform: • Functions — Farmer — Local assembler — Koperasi Unit Desa (KUD, village unit cooperative) — Non-KUD — National Logistic Agency (BULOG) — Private processor Wholesaler — Retailer Private household Farmer household Civil servant -- Hotel Restaurant Other services • Kinds of commodities Wet stalk paddy — Dry stalk paddy — Dry Gabah Menir (brewer rice) — Hull rice — Milled rice — Pounded rice Brown rice — Noodle rice Rice flour — Dry paddy • Kinds of activities — Irrigation — Land preparation — Seeding Fertilizing Planting — Cultivating — Protecting — Harvesting Commodity selling Field sorting Threshing Transporting Drying — Storing Sorting Purchasing — Product selling -- Hulling

- 65 - — Polishing Processing — Packaging Testing quality — Cooking Serving — Eating On the basis of these three categories and combinations of them, the Indonesian rice system can be characterized by five subsystems— producer, assembler, processor, wholesaler-retailer, and consumer—as indicated in Figure 1. The five subsystems in the flow model are linked by the exchange of commodity for money through the pricing mechanism or simply exchanges in commodity. The next step in the analytical flow model is to describe each subsystem by its major components. For each subsystem, a set of descriptors was established: 1. Producer subsystem 1.1 Irrigation type 1.2 Crop pattern 1.3 Technology level 2. Assembler subsystem 2.1 Commodity 2.2 Ownership 2.3 Integration 2.4 Technology level 3. Processor subsystem 3.1 Commodity 3.2 Ownership 3.3 Integration 3.4 Technology level 4. Wholesaler-retailer subsystem 4.1 Commodity 4.2 Ownership 4.3 Integration 4.4 Technology level 5. Consumer subsystem 5.1 Commodity 5.2 Location 5.3 Price 5.4 Number of people Each component can be modified by introducing additional subcom- ponents or deleting existing ones. For example, if a super high tech- nology is introduced, four levels of technologies would then be used

- 66 - D P PRODUCER 13 DODO ASSEMBLE? PRCXIESSOR J 1 WOLESALE DDD CONSUMER J FIGURE 1 Five subsystems of the Indonesian rice system flow model. Individual or group as actor in the subsystem (Q); kind of activities by its actors (O ); flow of commodity (-^); flow of money ( 4—). TABLE 1 Examples of the Producer and Assembler Subsystem Descriptions Producer Subsystem Description Assembler Subsystem Description Irrigation Type Crop Technology Pattern Level Owner- Commodity ship Integra- Technology tion Level Not irrigated Technical Rice- rice Rice- maize Low Medium High Low Medium High Rice Private Assemble Low only Medium High Private Pre- Low harvest Medium assemble High

- 67 - under technology level in the producer subsystem description (see Table 1 which shows examples of the producer and assembler subsystems). PERFORMANCE INDICATORS This section describes the performance indicators that measure the major components of each subsystem. A performance indicator should be distinguished from a goal or objective in that no specific level of achievement is predetermined. For this food system analysis, representative performance indicators are shown below for each subsystem: 1. Producer subsystem 1.1 Production 1.2 Employment 1.3 Net farm income 2. Assembler subsystem 2.1 Volume out/in 2.2 Employment 2.3 Net income 3. Processor subsystem 3.1 Volume out/in 3.2 Employment 3.3 Net income 4. Wholesaler-retailer subsystem 4.1 Volume out/in 4.2 Employment 4.3 Net income 5. Consumer subsystem 5.1 Daily consumption 5.2 Food expenditure 5.3 Percentage of total food expenditure spent on rice. This set of performance indicators represents the major measure of performance for each subsystem, and each subsystem description component is measured in terms of each performance indicator. The sum of the performance indicators for each component becomes the performance indicator for the total subsystem. New performance indicators can be added when necessary. A per- formance indicator is not valued in relation to another performance indicator. If the impact of a policy or technology improves the performance of one indicator and decreases that of another, the policy- maker must then make a relative evaluation of the indicators in order to make a decision regarding implementation of the policy. Subsystem performance indicators can be changed merely by changing from one component to another in the subsystem. Changes in performance

- 68 - indicators result from decision makers making decisions in response to external environmental factors such as changes in the technology level, type of irrigation used, etc. The model begins with the funds that the government expends via an institution on new programs or the development of new technologies and ends by projecting the impact of those funds on the designated performance indicator. The food system model provides a step-by-step approach to evaluating the impact of those expenditures (see Figure 2). The performance indicators can be expressed mathematically once data are supplied to the matrices and the flow of rice is defined. The descriptions, quantification, and linkage among institutions; decision criteria; and the action taken by decision makers to change the mix of subsystem components that affect the performance indicators cannot be evaluated by this flow model, however. Because this model will be used to assess the impact of government programs (policies) and the application of appropriate technologies on the Indonesian food system, the linking of institutions and decision criteria to the components of the subsystem and performance indicators is very important. COMPUTER PROGRAMMING The Food System Modeling System (MOD II)* is a set of programs and files designed to model an agricultural system. It divides the overall system into subsystems and formats each subsystem's data into reports. The original version used the Statistical Package for the Social Sciences (SPSS) and the SPSS Report Generator for this purpose, whereas the latest version utilizes reports written in Fortran. Each subsystem communicates with the others by means of three or four Fortran programs. These programs distribute the commodities output from one subsystem to the other subsystems. Another Fortran program reads input data and distributes them to separate files. Each Fortran program reports on the distribution of the data, and these are known as Audit Reports. Each subsystem within the model is designed to show the pertinent statistics for a portion of an agricultural system. The subsystems are: producer, assembler, processor, wholesaler-retailer, and consumer. Each subsystem is further broken down to unique processes which are described by unique combinations of key variables. Each line item within the subsystem reports is identified by unique combinations of four to six variables. For example, the assembler, processor, and wholesaler- retailer type subsystems use commodity code, ownership code, prime function code, integration code, and technical level code to describe uniquely each subsystem unit (or line item). The report programs can handle any number of these line items as long as each combination of the identifying variables is unique. The limitation on how many line items a subsystem can have is set by the limitations within the primary distribution program FTN1. To *This program was written in November 1979 and rewritten in April 1981. The author was Robert C. Hoppin, and it was revised by Tri Djoko Wahyono, Muhadi, Supriyanto, Kasiran, Hariadi Wardi.

- 69 - INSTITUTION AFFECTS THE DECISION CRITERIA THAT INFLUENCE THE I ,_ i r_,_^ MIX OF SUBSYSTEM COMPONENTS WHICH CHANGES THE r I .-_„_ PRODUCTION SUBSYSTEM DESCRIPTION ASSEMBLY SUBSYSTEM DESCRIPTION WHICH IS MEASURED BY THE PROCESSOR SUBSYSTEM DESCRIPTION PERFORMANCE INDICATOR r— i i _- "1 _-i FIGURE 2 A schematic of the general model.

- 70 - maximize the capacity of the system, the data distribution table is read from the Subsystem Transfer Matrix File, which is kept in a single table in the main memory. Given larger memory sizes, the table dimension may be increased for additional data distribution capacity. The current version of FTN1 will fit within 512K on the host processor. FTN1 and FTN5 have the same restriction on data variables, even though they require less space for their tasks. The limitations dictated by FTN1 in its current form are stated in terms of the limits placed on the number of distribution links in the distribution commodity mapping. Currently, there is room for 2,000 such links. As a basis for comparison, the largest number of links used in testing to date has been 127. Two thousand links seem more than sufficient for any conceivable use of the model. FTN1 currently provides for 30 different commodity groups, 10 commodity varieties, and 30 commodity forms within a subsystem. There is space for 30 each of the ownership codes, integration codes, and technical level codes. There is also room for 30 irrigation type codes. Within the data used on the system to date, these limitations seem more than sufficient. Note that space allocations have been set somewhat arbitrarily and should be adjusted (within the Job Control Language-JCL procedure files) if more space should be available and required. Blocking and other considerations of resource optimization were dictated by the host system and can be expected to change as experience is gained with the system. REFERENCE BPPT-SRI International. 1979. Indonesian Food System Manual Analytical Framework. BPPT, Jakarta, Indonesia.

SYSTEMS ANALYSIS AND A STUDY OF THE INDONESIAN FOOD SYSTEM Walter L. Fishel Assistant Director, Ohio Agricultural Research and Development Center, Ohio State University NEC Panelist Why systems analysis? The human mind is the best and most efficient assimilator and analyzer of relationships into useful knowledge; how- ever, whereas the capacity of the human mind is limited, the universe of knowledge is not. Systems analysis is nothing more than a conceptual approach to dealing with this problem by partitioning the universe into smaller and smaller sets of relationships, to a point at which the human mind can handle them. In other words, systems analysis is simply a method for organizing knowledge. Thus this methodology should never be used as a substitute for human judgment, but only as a way to facilitate its use in those situations in which human judgment is considered inadequate to deal with the complexity of the real world. In a very large sense, systems analysis is a mechanism for bringing intuition and analysis together. A second concept to keep in mind is that the end purpose of this activity is to provide policymakers with information that will help them make very important decisions. The systems analyst must always be oriented to the information of the decision maker and subordinate his interests as a professional. Incidentally, a systems analyst never asks a decision maker what information he wants; he must determine what information the decision maker needs. Thus interaction with the decision maker is very important. A systems analyst must work hard to maintain this information orientation, especially when he is struggling so hard to understand something as complex as the food system. CHARACTERISTICS OF INFORMATION Information has five basic dimensions: identification, precision, reliability, timeliness, and cost. Identification refers to how completely all of the relevant variables in the universe studied have been included in the system model. Precision refers to how accurately these variables are measured. Reliability refers to how well the information conforms to the decision maker's perception about that information. Timeliness refers to the availability of the information relative to when the decision maker needs the information. And, finally, cost refers to the cost in terms of money and staff effort required to generate the information. Reliability and timeliness are - 71 -

- 72 - the most important characteristics to the decision maker. Generally, and within reason, he is willing to sacrifice goodness in the other characteristics for an adequate performance by these two. Also, generally, this is exactly the opposite point of view of the profes- sional analyst in dealing with the same information. These trade-offs should be kept in mind. MODELING A FOOD SYSTEM The mechanisms of systems analysis have been developed around the world to the point that almost anything can be done analytically, given time and money. However, can a single systems analysis model that will respond to all the needs for information by decision makers be developed for an area as complex as the Indonesian food system? The obvious answer is that it might be possible but it is not practical. The food system is simply too complex in terms of its size, the number of com- ponents and variation in functions, the fact that human beings are involved at every level, the biological systems and weather, and the multidimensional values that are frequently exactly opposed to each other. Many implications of these attributes relate to how well one can structure one's knowledge about the food system, how much effort can be afforded to devote to data collection, and ultimately how difficult it will be to provide the information needed. To look more closely at two of these implications, first and most important it is necessary to take the time and effort to define the universe to be studied by determining at the start what questions the system model, once completed, should answer or what information it should provide. On the one hand, the complexity of the areas studied will limit the questions it will be possible to ask. On the other hand, determining which questions one wants to ask will in turn determine the size and scope of the structure of the system model and its data requirements. It is critical to the success of a systems analysis study to make this determination early in the study. This author's approach is to develop a paper that addresses the issues of problem definition, to review it with decision makers, and to change it based on these reviews. This also gives the systems analyst an extra chance to educate the decision maker, another of his major roles. Second, the analyst can give data too important a role in the development of system models. Especially in an area as complex as the food system, the analyst can become overwhelmed with data. Too frequently, analysts let data availability actually determine the system model, although the problem definition phase may indicate questions or information needs requiring a different model. Furthermore, because data collection is one of the less conceptual aspects of the study, the analyst may allow himself to become too taken up with acquiring it. For a system as complex as this one, it is often a waste of time to try to be too precise. A number of subjective estimation techniques can be used to obtain estimates from experts in each data area, and these would provide as good if not better estimates than published or collected data. This is the strength of partitioning in systems analysis.

- 73 - Finally, the systems analyst should not strive for perfection in the system model and data, because it is simply not possible. It is also not possible to expect to conceive of the entire universe all at once; this leads to frustration. The strength of systems analysis is its ability to reduce unmanageable universes to manageable components. In this, it is important to think "simple"—first partition and then integrate. A CAVEAT An outsider can be of only limited help in the development of a systems analysis study of the Indonesian food system and only at certain stages of the study, in particular, the stage during which one learns what it means to undertake a systems analysis. Another is assisting with the selection and application of appropriate analytical techniques. At other times, it may be best to send the outsiders away as they would probably do more harm than good. One implication of this is that Indo- nesian systems analysts should not seriously consider using a systems analysis model that has been developed for application elsewhere in the world. Their study must be unique if it is to be useful and enduring.