Irrigation and Drainage Systems 3: 229-240, 1989. © 1989 Kluwer Academic Publishers. Printed in the Netherlands.

The Juba Sugar Estate Game

RICHARD

C. CARTER

Silsoe College, Silsoe, Bedford, MK45 4DT, UK

Received 3 April 1989; accepted 6 April 1989

Key words: irrigation management, games, simulations, training

Abstract. An irrigation management training game is described. This game is based on the Juba Sugar Estate in Somalia and it involves the management of scarce inputs (water and fertiliser) and resources (labour, capital equipment, money and fuel). Participants take roles within small management teams and they make decisions concerning fertiliser application, irrigation, maintenance, harvest and cane haulage. The major objective of the game is to give the participants insights into the complex interactions between resources, inputs, activities and management decisions. A secondary aim is to encourage team work and mutual understanding of complementary job functions.

I. Introduction

Background

Role playing, g a m i n g a n d the use of s i m u l a t i o n s have long been recognised as v a l u a b l e l e a r n i n g techniques. As far back as the third c e n t u r y BC P l a t o extolled the virtues of play as a m e a n s of learning for life: • . . enforced learning will n o t stay in the m i n d . So avoid c o m p u l s i o n a n d let

your c h i l d r e n s ' lessons take the f o r m of play. (Plato, The R e p u b l i c ) A d u l t games a n d s i m u l a t i o n s have been used to train in warfare, business a n d c o m m e r c e and, m o r e recently, in agricultural m a n a g e m e n t (Brewer & Shubik 1979; E l g o o d 1981; Y o u n g m a n 1974). The advantages of games a n d simulations over other l e a r n i n g techniques are significant. Three particularly i m p o r t a n t ones are: -

The o p p o r t u n i t y for the s t u d e n t to e n c o u n t e r complex p r o b l e m s which require him or her to actively seek or develop p r o b l e m solving techniques. The possibility of testing decisions in a n e n v i r o n m e n t in which mistakes do

230 not matter (whereas in reality poor decisions may seriously affect peoples' lives or livelihoods or be very expensive); and perhaps most importantly in the present context. - The ability of a game to integrate techniques or disciplines which conventionally would be taught in isolated packages. Games therefore oppose the reductionist tendency of much analytical treatment of complex subjects. The Juba Sugar Estate Game (JSEG) is a recent addition to the growing list of participative training games and simulations for irrigation management staff. It has been influenced especially by the Green Revolution Game (Chapman 1982) and the Irrigation Game (Burton 1980). These two powerful teaching tools had been used frequently and successfully by the author prior to the start of work on the Juba Game. Despite its background the JSEG is now quite different to both of the Games just mentioned. It does however form a complementary member of this trio, putting over some particular lessons of its own.

For readers unfamiliar with the other two, the Green Revolution Game simulates the life of peasant farmers in South Bihar, India, and it raises, in particular, the issues associated with high yielding (high input-high output) rice varieties. The game is concerned with survival. The Irrigation Game is set on an Indonesian run-of-river irrigation scheme, where water management, crop selection and farmer/management interactions are important.

Development o f the game Work started on the JSEG in 1983, with the first version being completed the following year (Pole 1984). This groundwork benefitted greatly from Pole's first hand and detailed experience of the Estate itself. Development of the game continued in 1985-86, with significant modifications being introduced (Kenyon 1986). The present version has been consolidated by Chalinder (1987), and the game is now suitable for running with groups of 3 to 20 people. Development is continuing at the present time along the lines described below in Section 6.

The estate and the game The estate on which the game is modelled is owned by the Somali Government and run by Booker Agriculture International. It has been producing sugar since 1980. The most important features of the estate (which is described below in Section 2) are retained in the game, but inevitably certain simplifications have

231 been necessary. The most important aspect is that the game is modelling a commercial undertaking which must stay in profit. Both short term cash flow and medium to long term sustainability (through maintenance and capital replacement) are essential factors of the game. The participants' aim is the same as that of the estate management whose role they take: namely to maintain a steady flow of good quality cane to the factory.

2. The estate

Physical characteristics The estate is located next to the Juba River in Southern Somalia, about 120 km from the port of Kismayo (Fig. 1). It is situated on uneven land, on soils having a high degree of spatial variability. The total area of the estate is 8500 ha, although only one unit of 3500 ha is modelled in the game. One important feature of this unit of the estate is its elongated shape, with the factory at one end. This contributes to the problems of cane haulage following harvest. The estate is supplied with water from the perennial Juba River.

Agricultural practices Almost all the estate activities are carried out in a highly labour intensive manner. Planting of cane, weed control, fertiliser application, irrigation and cane cutting are all done by hand. The cane is a four to five year ratoon crop. Cultivations (disc ploughing) are mechanised, but this aspect is not modelled in the game. Harvesting (after burning) is by hand, with transport o f cane by tractors pulling double trailers. Harvesting is normally carried out after a 12 month growing season, with no irrigation in the last part of this period (drying off).

Irrigation Because of the variable soils and uneven topography most of the estate is irrigated from overhead. A conventional hand-move sprinkler system is used, with water being pumped from the main or branch canals to supply blocks of about 80 ha. The system is designed to apply 88 mm of water in an 11 hour set at peak, over a 7 day cycle of day and night irrigation. Pumping is by large multiple centrifugal units with diesel engines as the prime mover.

232

JUBA RIVER

FACTORY

5000m V~SCHEM E OCATION

~

~

Main \\ Pumping \~ Station ~

_FARM a 370 ha

\\ ~

3000m _FARM B 380 ha

1000m

I

1 FARM E 370 ha 1800m

2500m I 2000m

I

FARM D 300 ha

FARM C 530 ha

FARM F_ 350 ha 3000m FARM G 340 ha 1700m

FARM H 360 ha 1400m _FARM J 360 ha 1000m Note: FARM K 340 ha

Canal reach lengths not to scale

Overflow

Fig. 1. Location and schematic of the Juba Sugar Estate as simplified for the JSE Game.

Irrigation is carried out for ten months of the year (July to April) during which rainfall is insignificant. May and June are wet months.

233

Estate aims The aim of the estate is to produce a regular and high quality output of sugar. To achieve this the factory has to be supplied with cane in good condition in a continuous and even way over the harvest season (July to April).

Problems and constraints There are m a n y problems experienced in the running of the estate. These include: variable and unreliable river flows poor water quality at certain times of year fuel and fertiliser shortages - problems with spare parts - foreign exchange shortages - low sugar prices - cash flow problems

T h e

g a m e

Introduction and objectives The main aim of the JSE G a m e is to enable participants to experience the complex interactions between resources, inputs, activities and management decisions. In particular they are faced with decision making in situations of acute shortages of inputs and resources. A secondary and related aim is to encourage cooperation within a management team. The participants are divided into management teams, each consisting of three or four members. Each team is responsible for running the estate in a profitable manner and leaving it in a good state following play. Performance is judged in financial terms taking account of profits, maintenance status and provision for capital replacement. Also the cropping/harvest status of the estate is examined after play to assess the implications for future management (the elongated shape of the estate means that if several farms at the far end ripen simultaneously, there will be major cane haulage problems because of the distances involved). The game is divided into two-monthly playing periods and ideally a full season (five periods - the sixth being the wet season during which there is no harvest or irrigation) should be played. The two monthly periods correspond to individual growth stages of the crop. At the start of the game, the ten farms

234 are at different growth stages, just as in reality. The game is run by a Controller, ideally with an assistant if more than two groups are playing. The Controller has a small microcomputer to assist with calculations of canal losses and cane yields.

Activities, inputs and resources

The management teams have to make decisions about five m a j o r activities: fertiliser - harvest application - maintenance irrigation - cane haulage To carry out these activities the participants have access to inputs - water and fertiliser - and resources - labour, capital equipment, fuel and money. Any of these items, or several simultaneously, may be in short supply. There are several interactions between inputs and resources which influence the way in which management decisions are made. Fuel is used for irrigation pumps as well as the tractor fleet which hauls harvested cane. L a b o u r is required for fertiliser application, moving irrigation pipes, and for cane cutting. Fuel, fertiliser, labour and maintenance all have to be paid for out of cash surpluses generated from sugar production. Cane yield is limited by both fertiliser application rates and water availability. A simple model of yield reduction in relation to water stress is used by the Controller. The estate receives fuel and fertiliser deliveries every two months in the game, but the amounts delivered may not always match the requirements. -

-

Fertiliser application

Fertiliser is applied by hand during the growth stages 1 or 2 (or by split application in both stages). The chosen fertiliser rate limits m a x i m u m cane yields according to a production function given to the participants in graphical form. Labour requirements for fertiliser application vary with the amount applied and labour must be paid for.

Irrigation

At the beginning of each growth stage the participants are told how much water is available in the Juba River for abstraction at the main pumping station. Having calculated water requirements for each farm (from data on farm size, application efficiency, evapo-transpiration rates and growth stage) the team

235 then has to submit an indent for water for each farm. This must take account of estimated losses in the main and branch canals. The Controller runs a simple computer programme using a formula relating canal losses to discharge and reach length 1 and informs the team of the water actually available to each farm. Losses are calculated from the top end down the system and any shortages affect tail-end farms. The teams gradually learn to estimate losses more accurately, but they can also purchase a consultant's report from the Controller if they want to know the seepage loss formula which has been developed for the Juba system (see also Section 4). Water shortages affect yield of cane. Initially the teams are only informed of the relative sensitivity to stress of each of the growth stages. The Controller however has an empirical model on computer which relates stress in each stage to percentage yield reduction. The teams are told of the yield reductions due to water stress at the end of each stage, and this is presented in the form of a reduced maximum potential yield with which to enter the following stage. The modelling o f yield reductions due to water stress has to be simple and reasonably accurate. A set of multiplicative reduction factors has been derived using a procedure similar to that described by Tsakiris (1982). As with the canal seepage loss model, teams can purchase the table of yield reduction factors used, in the form of a consultant's report (see Section 4). Fuel and labour requirements for irrigation have to be calculated and paid for by the teams.

Harvest

In order to maintain the required flow of cane (1000-2000 t/day) to the factory two farms need to be harvested each period. Because of the elongated layout of the estate haulage difficulties arise if these two farms are both at the end distant from the factory. Cane cutting is done by hand and labour requirements have to be calculated and paid for. Haulage is done by the fleet of tractors pulling double trailers. Data are available concerning farm distances to the factory, tractor speeds, and fuel consumption. Fuel requirements have to be calculated and fuel paid for.

Maintenance

Maintenance of irrigation systems (pumps, pipelines and sprinklers) and of tractors is essential. As in reality maintenance can be neglected, deliberately or otherwise, but costly breakdowns become increasingly likely the longer maintenance is delayed.

236 In the case of tractors, working hours are logged and maintenance intervals are advised. The probability of breakdown increases rapidly once the recommended interval is exceeded and this probability is represented by sets of " c h a n c e " cards. Breakdown of pumping systems due to neglect of maintenance leads to water stress and possible yield reduction.

Capital replacement Irrigation systems and the tractor fleet have been assigned a nominal capital value and lifetime in the game. Money must be set aside for capital replacement and this is one of the criteria by which team performance is judged (see Section 5).

4. Special features of the game

Introduction The Juba Sugar Estate Game complements the other irrigation and agricultural management simulations which are available. However it has certain special features which to a greater or lesser extent distinguish it from the others, and these are described below.

Resource management and teamwork The major feature of the JSEG is the fact that irrigation is only one of many activities, resources and inputs which have to be managed in a coordinated way. Irrigation is not isolated from other cropping and management issues. Further, the interactions between the various aspects are an extremely important part of the game. Participants learn that decisions in one area, say irrigation, affect other activities, such as harvest or cane haulage. Management staff who may be used to working in rather isolated roles can begin to see how their decisions affect others.

The commercial aspect Many of the other available games very usefully focus on social issues which affect small farmers or their families. In addition some (for instance the

237 Irrigation Game, Burton 1980) draw attention to issues of communication between farmers and irrigation scheme management. The JSE Game however concentrates entirely on the commercial aim of producing a saleable commodity. This is clearly appropriate to the type of enterprise which is simulated and the game therefore usefully complements the others.

Introducing complexity and the use of consultant's reports One of the difficulties with games of any complexity is that a lot of time has to be spent prior to play explaining the rules and procedures. This can be tedious for the participants as well as using valuable time. In principle it is best to begin play as soon as possible using a simple version of the game and only introduce complexities as the participants become familiar with routine operations. One way in which complexity can be introduced is by the use of Consultants' Reports (this approach is not original to the present author, Youngman and Smith of Wye College having successfully used the technique previously). Two particular aspects of the JSE Game benefit from this phased approach. The first is the estimation of canal seepage losses. At the start of play the teams are simply told that the total scheme losses are usually around 2 5 - 3 0 % of water entering the main canal system. This is sufficient information to make rough estimates on which to base water indents. Later in the game the teams have the option of purchasing a consultant's report which presents an empirical equation relating losses to discharge and reach length. This is in fact the equation used by the Controller. Apart from the advantage of this approach outlined above, the teams begin to think about the monetary value of knowledge or scheme-specific information. The second aspect to which a consultant's report relates is that of cane yield reduction in response to water stress. Initially players are simply told of the sensitive growth stages o f the crop. Later they can purchase a report which presents the yield reduction factors used by the Controller. Other aspects of the game can benefit from this phased approach too, either in the way described or by using " c h a n c e " cards.

Operational flexibility Because of the small team approach used in the game it can be played with as few as 3, to as many as 20 participants. This is particularly useful in the development stages when frequent test runs are necessary, but it is useful too as an operational characteristic since group sizes are not always predictable.

238 5.

Conduct

of

the

Game

Introduction After an introduction to game-learning in general the Controller gives a short talk about the estate itself. This is illustrated with colourslides. He then explains as concisely as possible the mode of play, touching briefly upon some of the major issues. In its basic form the game can be played over shorter or extended time periods, but in both cases play is followed by discussion of major issues (arguably the most important activity) and by evaluation of team performance.

Game duration Like the other related games the JSE Game can be usefully played over a short intensive period (usually about 5 - 6 hours) as a self contained learning exercise. In this way it can contribute to long or short courses of education or training in irrigation or agricultural scheme management. In a long course (e.g., a taught postgraduate programme) the game could be extended over several weeks, giving more time for participants to plan their decisions. Ideally, in a short or long programme, the game should be integrated with related taught material on the various issues raised. This taught material could usefully include design exercises, so as to link design with management.

Post-play discussions It is important to synthesise the experiences of individuals and teams in a round-up discussion after play. Issues raised vary depending on the participants, but they often include some or all of the following: Factors influencing cane and sugar yield: water, nitrogen, fertiliser, pests and diseases, salinity, harvest management. Water scarcity: desirability of river regulation, need to estimate accurately and reduce canal losses, topend-tailend issue. - Cane haulage: problems of estate layout, tractor breakdowns, maintenance decisions. - Cash flow problems: needs of short-term recurrent funds and longer term capital replacement, foreign exchange scarcity, fluctuating sugar price. Management interactions: needs to coordinate decisions influencing various activities, difficulties of team work. -

239 - Management tools: value of knowledge and techniques in areas of yield prediction, calculation o f canal losses, decision making in situations of scarcity.

Performance evaluation After play (up to one season usually) the performance of each team is evaluated by taking account of three quantifiable factors and a fourth descriptive aspect. The first three factors are represented in cash terms and are: - cash profits made; - spending required to bring irrigation systems and tractor fleet back to fully maintained condition; and - extent of capital replacement investment. The fourth aspect is the cropping situation and the extent to which poorly planned harvesting will have contributed to future management problems. This is particularly related to the layout of the Juba Estate.

6. Future d e v e l o p m e n t o f the G a m e

As.the game stands the Controller makes limited use of a microcomputer for checking important calculations. Ideally far more of these calculations should be carried out on the computer so that all team decisions are independently checked. A further development would be to set up the whole game on a microcomputer so that a single player could run it. This would still achieve the learning objective of exposing the player to resource and input interactions, but the element o f personal interactions within a management team would disappear. Post-play discussion would be less meaningful too, since much of the value of this depends on the sharing of a variety of experiences. A more radical development in the same direction would be to turn the game into a more realistic simulation and thereby create a management tool. To achieve this a substantial amount of work would be needed to upgrade the model of yield response to water and fertiliser. In addition the whole simulation would need to be programmed for microcomputer use and management techniques such as linear programming would be introduced.

Note

1. q = 0.00125Q 0.5. L where q = losses (l/s), Q = discharge (l/s) and L = reach length (m) (an adaptation of the Moritz formula, quoted in Dhillon 1967).

240

References Brewer G.D. & Shubik M. 1979. A Critique of Military Problem Solving. Harvard University Press. Burton M.A. 1980. Irrigation Management Training Manual. Sir M. MacDonald and Partners, Consulting Engineers, Cambridge, UK. Chalinder P.A. 1987. The Juba sugar estate simulation game. Unpublished MSc thesis, Silsoe College, Cranfield Institute of Technology, UK. Chapman G.P. 1982. The Green Revolution Game. Production by Marginal Context Ltd, 36 St Andrew's Rd, Cambridge, UK. Dhillon G.S. 1967. Estimation of seepage losses from lined channels. Indian Journal of Power and River Valley Development 17: 16-20. Elgood C. 1981. Handbook of Management Games. Gower Publishing. Kenyon A. 1986. The Juba sugar project simulation - a training exercise for irrigation management. Unpublished MSc thesis, Silsoe College, Cranfield Institute of Technology, UK. Plato (c. 360 BC) The Republic. Pole M.B.F. 1984. A simulation exercise for use in the training of irrigation managers. Unpublished MSc thesis, Silsoe College, Cranfield Institute of Technology, UK. Tsakiris G.P. 1982. A method for applying crop sensitivity factors in irrigation scheduling. Agricultural Water Management 5: 335-343. Youngman J.P. 1974. Construction, implementation and appraisal of a farm management game. Unpublished M. Phil thesis, Wye College, University of London.