package examples;

/* --------------------------------------------------------------------------

* File: Diet.java

* Version 12.9.0

* --------------------------------------------------------------------------

* Licensed Materials - Property of IBM

* 5725-A06 5725-A29 5724-Y48 5724-Y49 5724-Y54 5724-Y55 5655-Y21

* Copyright IBM Corporation 2001, 2019. All Rights Reserved.

*

* US Government Users Restricted Rights - Use, duplication or

* disclosure restricted by GSA ADP Schedule Contract with

* IBM Corp.

* --------------------------------------------------------------------------

*

* Reads data for a dietary problem, builds the model, and solves it.

*

* Input data:

* foodMin[j] minimum amount of food j to use

* foodMax[j] maximum amount of food j to use

* foodCost[j] cost for one unit of food j

* nutrMin[i] minimum amount of nutrient i

* nutrMax[i] maximum amount of nutrient i

* nutrPerFood[i][j] nutrition amount of nutrient i in food j

*

* Modeling variables:

* buy[j] amount of food j to purchase

*

* Objective:

* minimize sum(j) buy[j] * foodCost[j]

*

* Constraints:

* forall foods i: nutrMin[i] <= sum(j) buy[j] * nutrPer[i][j] <= nutrMax[j]

*

* Options:

* - build the model by column, rather than by row

* - restrict to selecting integral food quantities

* - create a multi-objective model

*

* See the usage message for more details.

*/

import ilog.concert.*;

import ilog.cplex.*;

import ilog.cplex.IloCplex.MultiObjIntInfo;

import ilog.cplex.IloCplex.MultiObjLongInfo;

import ilog.cplex.IloCplex.MultiObjNumInfo;

public class Diet {

/**

* Print a usage message to stderr and abort.

*/

static void usage() {

System.err.println();

System.err.println("Usage: Diet [options...] <datafile>");

System.err.println(" By default, the problem will be generated by rows.");

System.err.println();

System.err.println(" Supported options are:");

System.err.println(" -r generate problem by row");

System.err.println(" -c generate problem by column");

System.err.println(" -i restrict to integral food quantities (i.e., convert to a MIP)");

System.err.println(" -m create a multi-objective problem where, in addition to");

System.err.println(" minimizing cost, the number of unique foods selected is");

System.err.println(" maximized (can be combined with -i)");

System.err.println();

System.exit(2);

}

/**

* A class to store problem data.

*/

static class Data {

int nFoods;

int nNutrs;

double[] foodCost;

double[] foodMin;

double[] foodMax;

double[] nutrMin;

double[] nutrMax;

double[][] nutrPerFood;

Data(String filename) throws IloException, java.io.IOException, InputDataReader.InputDataReaderException {

InputDataReader reader = new InputDataReader(filename);

foodCost = reader.readDoubleArray();

foodMin = reader.readDoubleArray();

foodMax = reader.readDoubleArray();

nutrMin = reader.readDoubleArray();

nutrMax = reader.readDoubleArray();

nutrPerFood = reader.readDoubleArrayArray();

nFoods = foodMax.length;

nNutrs = nutrMax.length;

if (nFoods != foodMin.length || nFoods != foodMax.length)

throw new IloException("inconsistent data in file " + filename);

if (nNutrs != nutrMin.length || nNutrs != nutrPerFood.length)

throw new IloException("inconsistent data in file " + filename);

for (int i = 0; i < nNutrs; ++i) {

if (nutrPerFood[i].length != nFoods)

throw new IloException("inconsistent data in file " + filename);

}

}

}

static void buildModelByRow(IloModeler model, Data data, IloNumVar[] buy, IloObjective cost, IloNumVarType type)

throws IloException {

int nFoods = data.nFoods;

int nNutrs = data.nNutrs;

for (int j = 0; j < nFoods; j++) {

buy[j] = model.numVar(data.foodMin[j], data.foodMax[j], type);

}

cost.setExpr(model.scalProd(data.foodCost, buy));

for (int i = 0; i < nNutrs; i++) {

model.addRange(data.nutrMin[i], model.scalProd(data.nutrPerFood[i], buy), data.nutrMax[i]);

}

}

/**

* 按列的形式添加变量 .<br>

*/

static void buildModelByColumn(IloMPModeler model, Data data, IloNumVar[] buy, IloObjective cost,

IloNumVarType type) throws IloException {

int nFoods = data.nFoods;

int nNutrs = data.nNutrs;

IloRange[] constraint = new IloRange[nNutrs];

for (int i = 0; i < nNutrs; i++) {

constraint[i] = model.addRange(data.nutrMin[i], data.nutrMax[i]);

}

for (int j = 0; j < nFoods; j++) {

IloColumn col = model.column(cost, data.foodCost[j]);

for (int i = 0; i < nNutrs; i++) {

col = col.and(model.column(constraint[i], data.nutrPerFood[i][j]));

}

buy[j] = model.numVar(col, data.foodMin[j], data.foodMax[j], type);

}

}

static void createMultiObj(IloCplex cplex, IloNumVar[] buy, IloObjective cost, IloObjective variety)

throws IloException {

// Create binary variables for each food.

IloNumVar[] varUsed = cplex.numVarArray(buy.length, 0.0, 1.0, IloNumVarType.Bool);

// Add indicator constraints that force the binary variables to 1 if

// a food is purchased/used.

for (int i = 0; i < buy.length; ++i) {

cplex.add(cplex.ifThen(cplex.eq(varUsed[i], 0), cplex.eq(buy[i], 0.0)));

cplex.add(cplex.ifThen(cplex.eq(buy[i], 0.0), cplex.eq(varUsed[i], 0)));

}

// Set up the second objective so that it maximizes the sum of the

// binary variables.

variety.setExpr(cplex.sum(varUsed));

variety.setSense(IloObjectiveSense.Maximize);

// Create an array of objectives:

// - the first objective minimizes the cost

// - the second objective maximizes the variety

IloNumExpr[] objArray = new IloNumExpr[] { cost.getExpr(), variety.getExpr() };

// Use a negative weight to maximize the second objective. The

// objective senses of all objective functions must match. Hence, we

// have to formulate max(sum) as min(-sum).

double[] weights = new double[] { 1.0, -1.0 };

// Adjust priorities such that we optimize the first objective

// first (i.e., give it a higher priority).

int[] priorities = new int[] { 2, 1 };

// Allow a small degradation in the first objective.

double[] absTols = new double[] { 0.5, 0.0 };

double[] relTols = new double[] { 0.0, 0.0 };

// 多目标设置

// Add the lexicographic objective to the model.

cplex.add(cplex.minimize(cplex.staticLex(objArray, weights, priorities, absTols, relTols, "staticLex1")));

}

static void printSolutionStat(String what, String val) {

System.out.printf(" %s: %s%n", what, (val != null) ? val : "---");

}

static void printMultiObjInfo(IloCplex cplex, MultiObjIntInfo what, int subprob) throws IloException {

String valString = null;

try {

valString = Integer.toString(cplex.getMultiObjInfo(what, subprob));

} catch (IloException ignore) {

}

printSolutionStat(what.toString(), valString);

}

static void printMultiObjInfo(IloCplex cplex, MultiObjLongInfo what, int subprob) throws IloException {

String valString = null;

try {

valString = Long.toString(cplex.getMultiObjInfo(what, subprob));

} catch (IloException ignore) {

}

printSolutionStat(what.toString(), valString);

}

static void printMultiObjInfo(IloCplex cplex, MultiObjNumInfo what, int subprob) throws IloException {

String valString = null;

try {

valString = Double.toString(cplex.getMultiObjInfo(what, subprob));

} catch (IloException ignore) {

}

printSolutionStat(what.toString(), valString);

}

static void printSolutionStats(IloCplex cplex) throws IloException {

final int num = cplex.getMultiObjNsolves();

for (int i = 0; i < num; ++i) {

int prio = cplex.getMultiObjInfo(MultiObjIntInfo.MultiObjPriority, i);

System.out.printf("subproblem %d [priority %d]:%n", i, prio);

printMultiObjInfo(cplex, MultiObjIntInfo.MultiObjError, i);

printMultiObjInfo(cplex, MultiObjIntInfo.MultiObjStatus, i);

printMultiObjInfo(cplex, MultiObjNumInfo.MultiObjTime, i);

printMultiObjInfo(cplex, MultiObjNumInfo.MultiObjDetTime, i);

printMultiObjInfo(cplex, MultiObjLongInfo.MultiObjNiterations, i);

printMultiObjInfo(cplex, MultiObjNumInfo.MultiObjObjValue, i);

printMultiObjInfo(cplex, MultiObjNumInfo.MultiObjBestObjValue, i);

printMultiObjInfo(cplex, MultiObjLongInfo.MultiObjNnodes, i);

printMultiObjInfo(cplex, MultiObjLongInfo.MultiObjNnodesLeft, i);

}

}

public static void main(String[] args) throws Exception {

// Set default arguments and parse command line.

String filename = "./data/diet.dat";

boolean byColumn = false;

IloNumVarType varType = IloNumVarType.Float;

boolean multiobj = false;

for (int i = 0; i < args.length; i++) {

if (args[i].charAt(0) == '-') {

if (args[i].equals("-r"))

byColumn = false;

else if (args[i].equals("-c"))

byColumn = true;

else if (args[i].equals("-i"))

varType = IloNumVarType.Int;

else if (args[i].equals("-m"))

multiobj = true;

else {

System.err.printf("Unknown argument: %s%n", args[i]);

usage();

}

} else {

filename = args[i];

}

}

// Read the data.

Data data = new Data(filename);

final int nFoods = data.nFoods;

IloCplex cplex = new IloCplex();

try {

// Build model

IloNumVar[] buy = new IloNumVar[nFoods];

IloObjective cost = cplex.minimize();

cost.setName("cost");

if (byColumn)

buildModelByColumn(cplex, data, buy, cost, varType);

else

buildModelByRow(cplex, data, buy, cost, varType);

// Optionally, create a second objective.

IloObjective variety = cplex.maximize();

variety.setName("number of foods");

if (multiobj) {

createMultiObj(cplex, buy, cost, variety);

} else {

cplex.add(cost);

}

// Write a copy of the problem to a file.

cplex.exportModel("Diet.java.lp");

// Solve model

if (multiobj) {

// Set multi-objective display level to "detailed".

cplex.setParam(IloCplex.Param.MultiObjective.Display, 2);

// Purely for demonstrative purposes, set global and local limits

// using parameter sets.

// First, set the global deterministic time limit.

cplex.setParam(IloCplex.Param.DetTimeLimit, 60000);

// Second, create a parameter set for each priority.

IloCplex.ParameterSet[] params = new IloCplex.ParameterSet[2];

params[0] = new IloCplex.ParameterSet();

params[1] = new IloCplex.ParameterSet();

// Set the local deterministic time limits. Optimization will stop

// whenever either the global or local limit is exceeded.

params[0].setParam(IloCplex.Param.DetTimeLimit, 50000);

params[1].setParam(IloCplex.Param.DetTimeLimit, 25000);

// Optimize the multi-objective problem and apply the parameter

// sets that were created above. The parameter sets are used

// one-by-one by each optimization.

if (!cplex.solve(params))

throw new IloException("Failed to optimize");

} else {

if (!cplex.solve())

throw new IloException("Failed to optimize");

}

// Print the solution status.

System.out.println();

System.out.println("Solution status = " + cplex.getStatus());

// Print the objective value(s).

if (multiobj) {

System.out.printf("Solution value 0 (%s) = %f%n", cost.getName(), cplex.getValue(cost.getExpr(), -1));

System.out.printf("Solution value 1 (%s) = %f%n", variety.getName(),

cplex.getValue(variety.getExpr(), -1));

System.out.println();

printSolutionStats(cplex);

} else {

System.out.printf("Solution value = %f%n", cplex.getObjValue());

}

System.out.println();

// Print the solution values.

for (int i = 0; i < nFoods; i++) {

System.out.printf("Food %d: Buy = %f%n", i, cplex.getValue(buy[i]));

}

} finally {

cplex.end();

}

}

}