AU2010202862B1

AU2010202862B1 - Banana production prediction and mangement

Info

Publication number: AU2010202862B1
Application number: AU2010202862A
Authority: AU
Inventors: Robert Anton Breinl; Paul Anthony EDWARDS
Original assignee: P Edwards Invest Pty Ltd
Current assignee: P EDWARDS INVESTMENTS Pty Ltd
Priority date: 2010-07-07
Filing date: 2010-07-07
Publication date: 2010-11-11
Anticipated expiration: 2030-07-07

Description

AUSTRALIA PATENTS ACT 1990 Complete Patent Specification BANANA PRODUCTION PREDICTION AND MANAGEMENT The following statement is a full description of the best method of performing this invention: 1 BANANA PRODUCTION PREDICTION AND MANAGEMENT TECHNICAL FIELD 5 The present invention relates to methods for crop prediction and management. Particular embodiments of the invention are concerned with banana crops. BACKGROUND 10 The reference to any prior art in the following discussion is not to be interpreted as any admission, or evidence, that such prior art formed or forms part of the common general knowledge. Banana farming is performed throughout many tropical and subtropical 15 regions of the world. It is particularly popular in northern Queensland Australia. Bananas are a perennial crop. They are planted as bits of banana corm or as tissue cultured plantlets. As the plant tree grows it develops suckers, 1 or 2 of 20 which are selected to grow as the following crop ( 1st ratoon ). The bunch of the plant tree is harvested. The selected suckers grow (1ist ratoon) and in turn develop suckers which are selected ( 2 nd ratoon) as well as a bunch which is harvested. 25 The selected suckers grow as the 2 nd ratoon and so on. Plant and 1 st ratoon crops are generally even (i.e. they are synchronous in that they bunch at the same time) and consequently are the most efficient to farm. 30 However, this synchronicity is quickly lost as the ratoons evolve, until the farmer has a totally asynchronous crop which requires farming every week of the year. In such a situation the crop effectively manages the farmer. These 2 blocks are the most inefficient to farm, but comprise the majority of blocks farmed in North Queensland. It is an object of the present invention to address the above described 5 disadvantages of the prior art. SUMMARY OF THE INVENTION According to a first aspect of the present invention, there is provided a method for managing a banana crop on a farm including the steps of: 10 o capturing field data for the crop; o manipulating the field data with a computer to produce current belling sequence predictions for a subsequent period; o determining a desired belling sequence for the crop; o making a comparison of the current belling sequence predictions 15 for the subsequent period with the desired belling sequence; o and based on said comparison implementing strategies on the farm to achieve the desired belling sequence. Preferably the step of capturing field data for the crop includes performing a 20 stool count and preferably carrying out a measuring survey block by block across the farm. According to a further aspect of the invention, there is provided a computer software program for execution by an electronic computer for managing a 25 banana crop on a farm, said software program containing instructions to: o receive field data for the crop; o manipulate the field data to produce current belling sequence predictions for a subsequent period; o receive a desired belling sequence for the crop; 30 o make a comparison of the current belling sequence predictions for the subsequent period with the desired belling sequence; o display a report for the comparison.

3 The method may include a step of predicting the size and timing of the crop a predetermined period in advance. Preferably the method includes calculating a carton number for the predicted 5 crop size. In a preferred embodiment of the invention the method also includes the steps of monitoring the effect of said strategies and determining if updated belling sequence predictions are approaching the desired belling sequence. 10 The method may further include a step of altering the strategies in the event of the updated belling sequence predictions not approaching the desired belling sequence. 15 A webserver may be provided that is loaded with the software and arranged to provide online access to the software functionality to a plurality of users online. BRIEF DESCRIPTION OF THE DRAWINGS 20 Preferred features, embodiments and variations of the invention may be discerned from the following Detailed Description which provides sufficient information for those skilled in the art to perform the invention. 25 The Detailed Description is not to be regarded as limiting the scope of the preceding Summary of the Invention in any way. The Detailed Description will make reference to a number of drawings as follows: Figure 1 is a graph showing the relationship between leaf number and 30 tree height of banana trees that is used in a software product according to a preferred embodiment of the present invention. Figure 2 is the main flowchart of the method according to a preferred embodiment of the present invention.

4 Figure 3 is a flowchart of the backward target belling prediction method according to a preferred embodiment of the present invention. Figure 4 is a flowchart of the bunch ripening prediction method according 5 to a preferred embodiment of the present invention. Figure 5 is a view of the main user interface for a computer implemented version of the method of figure 2. 10 Figure 6 is a view of a target belling prediction user interface including graph output for a computer implemented version of the method of figure 3. Figure 7 is a view of a bunch ripening prediction user interface 15 including graph output for a computer implemented version of the method of figure 4. Figure 8 is an eveness prediction table generated by a method according to a preferred embodiment of the present invention. 20 Figure 9 is a banana sucker tree analysis graph generated by a method according to a preferred embodiment of the present invention. Figure 10 is a belling prediction graph generated by a method according 25 to a preferred embodiment of the present invention. Figure 11 is a table guide to belling generated by a method according to a preferred embodiment of the present invention. 30 Figure 12 is a computer user interface to extract historical growth rate data by a method according to a preferred embodiment of the present invention. Figure 13 is a picture of taking a banana tree height measurement.

5 Figure 14 is a graph relating ratoon cycles to weeks after planting. Figure 15 is a computer user interface to validate and format input field 5 data by a method according to a preferred embodiment of the present invention. DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS 10 Part 1. Embodiments of the present invention promote predicted and controlled production outcomes for banana farms. The invention assists in producing, 15 even, i.e. synchronous, cropping blocks, with each block made to produce sequentially. The method relies upon a model developed through field trials which provides the following key relationships within banana tree growth as follows: 20 1. That banana trees produce bell (flower / bunch) at a constant and predictable leaf number. 2. That a constant and reliable relationship exists between the tree height and leaf number. 25 In relation to point 2. above, figure 1 is a scatter plot graph showing the relationship between tree height to leaf number from field trial data. This graph represents several years research and data collection to obtain the relationship illustrated therein. Tree leaf number and height relationships 30 were determined for a range of tree sizes and for trees of different ratoon cycles. These relationships allow for a simple method, tree height measurements to be used first to assess a banana paddock and also to apply any management practices to that paddock.

6 The field trials also measured growth following the application of nurse sucker pruning treatment, i.e. cutting out the mother tree before maturity, sacrificing the mother, to give the following child sucker a growth spurt and bring forward eventual fruit yields. Again the trials showed constant and predictable height 5 to bell leaf number relationships which could be used to predict cropping results after the application of nurse sucker techniques. Software implementing the preferred method uses the key field trial results to process field tree height data to make accurate belling and crop production 10 predictions. The software analyses the tree height field data input and produces an output profile of the numbers and distributions of the sucker trees present in the paddock at the time of sampling. The software also then simulates future tree growth in a banana paddock, based on historical growth rates measured in the course of previous crop monitor activities. 15 A measured sample of trees from a given paddock is "grown" in silico to belling maturity. Computer output produces graphs, lists and tables of monthly belling forecasts, timing and numbers. The software allows for tracking the progress of the crop throughout the year, to reconcile predicted 20 with actual belling. The belling forecasts for each paddock measured are combined in custom designed spreadsheets to give a prediction of whole farm production. Part 2 25 The following sequence of activities are utilised in applying a method according to an embodiment of the invention to a new client farm. The sequence of events for introducing the crop management service to client 30 is as follows: 1. Initial assessment - discuss with client their desired production outcomes.

7 2. Stool count and carry out measuring survey across farm. For a 300 hectare farm this step usually takes six persons three weeks to perform depending on weather conditions. 3. Input and manipulate field data to produce current belling sequence 5 predictions for the next 15 months. 4. Data output from software and produce spreadsheets. 5. Use carton to bell software to generate desired belling sequence. 6. Compare with desired belling and investigate possible strategies to achieve it. 10 7. Report findings to client. Discuss managment options arriving at a management plan. 8. Provide extension to facilitate plan or provide expert labour to carry it out on behalf of client. 9. Monitor monthly to reconcile growth actual and belling sequence 15 compared with predicted growth and belling sequence. 10. Provide any advice required to manage production. 11. Reassess and alter strategy as required. Therefore a farmer might employ the method to predict the progressive 20 harvesting best practice for his farm, i.e. as one block's plants are ready to be bagged or harvested another adjoining could be ready for bagging or harvest for continuous productivity. It is also possible to delineate what activity is needed by the farmer to achieve best practice. 25 Part 3 The computer software receives the input and generates output. Computer input. 30 The inputs to the software package include: o Collected field sample measurements of upcoming sucker tree heights o Tree growth rates (leaf emergence) over one year 8 o Average banana tree belling height o Belling leaf number o Block stool counts o Sampling date 5 o Weekly belling counts o Bell to maturity bunch hang times over one year o Carton rates According to a preferred embodiment of the invention, a computer assisted 10 method is provided that is illustrated in the figure 2 flowchart boxes and figure 5 computer user interface. Red-font numerals in figure 5 refer to figure 2 flowchart box numbering. The steps that take place at the various boxes in the figure 2 flowchart boxes are as follows: 15 A 5% sample of banana tree heights growing in a paddock block is measured. The sucker height is measured from ground to throat (box 4), (fig 13). Input data is automatically checked and validated before submission for analyses and saving (boxes 5, 6) (fig 15). 20 Other inputs include historical growth rates (boxes 2, 3) extracted from normal crop monitoring activities (fig 12). Other inputs also include parameters and settings for the particular block 25 including name, stool count, sample date, bell height, bell leaf number (box 1). Bunch ripening hang times, carton rates (cartons per bunch) and new bell flower counts is input for bunch ripening and carton number prediction (boxes 15, 17), (figs 3, 4 and 6, 7). 30 All inputs are collated and stored as jobsheets (fig 5) which are HTML documents with included JavaScripts to process the input data and facilitate the various outcomes discussed herein.

9 Computer output Outcomes can be presented as graphs, tables and lists or linked combinations 5 of all, which can be exported for use in spreadsheets where necessary. Output from different processes is described in flowchart boxes 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17 and some is illustrated in figures 5, 7, 8, 9, 10, 11, 15. 10 The compiled data allows an accurate prediction to be made of the time when the bananas would be belled (flower) up to 15 months hence. The outcome may be predicted for normal growth (box 12), (fig 10), (fig 5 - 12) or alternately if nurse sucker or various sucker size selection strategies were employed (boxes 10, 11), (fig 8). 15 Output includes: o Analysis of sucker sizes and distributions (box 7), (fig 9) o Tree height and leaf number relationship calculator (box 8) o Table of mother - sucker height relationships (box 9) 20 o Belling date guide table (box 13), (fig 11) o Spreadsheet documents. Results are presented to farmers as spreadsheets and related tables graphs and lists of bell emergence predictions (boxes 16), (figures 8, 9, 10, 11). o A bunch ripening hang time module, (boxes 15), (figs 4, 7) generates 25 output that predicts expected patterns of mature bunches and carton numbers from new bell flower input. Actual belling data from farms can be input to further reconcile prediction and production. o A bunch ripening hang time module, (box 17), (figs 3, 6) makes a theoretical backward prediction of the belling pattern required to fulfil a 30 desired carton production. In addition crop monitoring, tracking (box 14) provides a system of monitoring as the plants grow to ensure the targeted outputs are on track according to 10 the data and calculations already accumulated. A monthly review is conducted, to ensure the whole process is accurate enough to predict in future seasons. 5 Data is stored in job folders which contain simple HTML jobsheet files and spreadsheets. One jobsheet contains the input from one block (paddock). The folders also contain hotlinks to online bunch hang ripening prediction software. Part 4. 10 In one aspect of the invention there is provided a software product that implements the methods illustrated in boxes in flowchart figure 2 and computer user interface figure 5 and also flowcharts figures 3, 4 and computer user interfaces figures 6, 7. 15 Figure 2 shows the major steps in a method according to a preferred embodiment of the invention to analyse upcoming sucker tree heights and to simulate crop growth from sucker to belling stage and produce output reports. Steps with red-font numerals in figure 2 boxes link to red-font numerals 20 overlayed on computer user interface figure 5. The product's core software functions from computer user interface figure 5 include: 25 1. A functions to extract growth rate (leaf emergence) data for historical spreadsheet records (boxes 2 , 3) (figs 5 - 2, 3), (fig 12). 2. A function to validate and format field data input, typographic error detection (fig 15), (boxes 5), (figs 5 - 5). 3. A function which updates the HTML web page jobsheet in the browser 30 for saving while offline processing (box 6), (fig 5 - 6). 4. Sucker height and distribution profile analysis for the sampled paddock (box 7), (fig 5 - 7), (fig 9).

11 5. In silico growth simulation functions (box 10, 11, 12), (fig 5 - 10, 12, 13). 6. A function which simulates paddock growth if suckers within a certain height range (even cropping) are selected as followers during normal 5 desucker procedures (box 11, 12), (fig 5 - 11), (figs 10, 8). 7. A function which simulates paddock growth if nurse sucker pruning is applied (box 10) to the trees in the paddock including the cut off height of the mother trees to be sacrificed (fig 5 - 10). This function also calculates the percentage of trees which will be cut during the nurse 10 sucker pruning application. 8. A function to predict bell timing of a fixed range of tree heights (box 13), (fig 5 - 13), (fig 11). This is a guide tool, a guide predicting which trees of a certain height will bell in which month. 9. A function which calculate how many leaves will have been in a tree of 15 a certain height and also what will be the height of a tree that has had a certain number of leaves (box 8), (fig 5 - 8). 10.A function which calculates the height relationships between mothers and suckers (box 9), (fig 5 - 9). 11.A module to predict subsequent ratoon bell time arrivals after a fixed 20 planting date (fig 14). 12.A module of heat unit analysis of climate data to assess the potential of new growing areas. The product's additional software functions from computer user interfaces 25 figures 6 and 7 include: 1. A planning module which uses bunch hang time data to assess possible belling pattern and numbers required for subsequent desired fruit maturity and carton per month production (box 17), (fig 3, 6).

12 2. An online module which uses bunch hang time data to predict cartons per month production from weekly new bell counts and carton ratios. Predictions can be compared with actual carton production (box 15), (fig 4, 7) 5 The software product may be implemented in a spreadsheet such as an Excel spreadsheet. We have a collection of spreadsheet document templates to compile computer generated belling output and bunch fill time and carton ratio data to make full and final production predictions in cartons 10 per month. Part 5 A software product to implement the invention, according to one embodiment, 15 includes instructions for a computer to receive field data for the crop, manipulate the field data to produce current belling sequence predictions for a subsequent period, make a comparison of the current belling sequence predictions for the subsequent period with the desired belling sequence and display a report for the comparison. 20 The software product may be provided on a machine readable media such as an optical or magnetic disk or integrated circuit memory. Furthermore, the software product may be loaded onto a web server arranged to provide online access to the software functionality to a plurality of users online. 25 In compliance with the statute, the invention has been described in language more or less specific to structural or methodical features. The term "comprises" and its variations, such as "comprising" and "comprised of" is used throughout in an inclusive sense and not to the exclusion of any 30 additional features. It is to be understood that the invention is not limited to specific features shown or described since the means herein described comprises preferred forms of putting the invention into effect. The invention is, 13 therefore, claimed in any of its forms or modifications within the proper scope of the appended claims appropriately interpreted by those skilled in the art.

Claims

1. A method for managing a banana crop on a farm including the steps of capturing field data for the crop; manipulating the field data with a computer to produce current belling sequence predictions for a subsequent period; determining a desired belling sequence for the crop; making a comparison of the current belling sequence predictions for the subsequent period with the desired belling sequence; and based on said comparison implementing strategies on the farm to achieve the desired belling sequence.

2. A method according to claim 1, wherein the step of capturing field data for the crop includes performing a stool count.

3. A method according to claim 1 or claim 2, wherein the step of capturing field date for the crop includes carrying out a measuring survey across the farm.

4. A method according to any one of the preceding claims including a step of predicting the crop belling sequence a predetermined period in advance.

5. A method according to any one of the preceding claims, further including the steps of monitoring the effect of said strategies to determine if updated belling sequence predictions are approaching the desired belling sequence.

6. A method according to claim 4, including altering the strategies in the event of the updated belling sequence predictions not approaching the desired belling sequence

7. A method according to claim 4, including calculating a carton number corresponding to predicted belling sequence. 15

8. A computer software program for execution by an electronic computer for managing a banana crop on a farm, said software program containing instructions to: receive field data for the crop; manipulate the field data to produce current belling sequence predictions for a subsequent period; receive a desired belling sequence for the crop; make a comparison of the current belling sequence predictions for the subsequent period with the desired belling sequence; and display a report for the comparison.

9. A web server loaded with the computer software program of claim 8 and arranged to provide online access to the software functionality to a plurality of users online.

10. A method substantially as described herein with reference to the figures. * * *