算法优化

src/main/java/com/aps/entity/Algorithm/Chromosome.java

@@ -23,7 +23,7 @@ public class Chromosome {
 
     private String ID = UUID.randomUUID().toString();
 
-
+    private String generateType;
 
     private Integer version=0;
 
@@ -117,6 +117,12 @@ public class Chromosome {
     */
     private double[] WeightedObjectives;//越靠近1越优
     private double WeightedObjective =0; // 加权目标值（用于自定义权重）
+
+    /// <summary>
+    /// 适应度值
+    /// </summary>
+    private double[] fitnessLevel;
+
     /// <summary>
     /// 适应度值
     /// </summary>

src/main/java/com/aps/entity/basic/GlobalParam.java

@@ -2,10 +2,7 @@ package com.aps.entity.basic;
 
 import lombok.Data;
 
-import java.util.ArrayList;
-import java.util.HashMap;
-import java.util.List;
-import java.util.Map;
+import java.util.*;
 
 /**
  * 作者：佟礼
@@ -81,6 +78,8 @@ public class GlobalParam {
         objectiveConfigs.add(new ObjectiveConfig(OBJECTIVE_SETUP_TIME, true, 2, 0.1));
         objectiveConfigs.add(new ObjectiveConfig(OBJECTIVE_MACHINE_LOAD, false, 3, 0.1));
         objectiveConfigs.add(new ObjectiveConfig(OBJECTIVE_TARDINESS, true, 1, 0.3));
+
+        objectiveConfigs.sort(Comparator.comparing((ObjectiveConfig op) -> op.getLevel()));
     }
 
     /// <summary>
@@ -146,6 +145,8 @@ public class GlobalParam {
     /// </summary>
     public void addObjectiveConfig(ObjectiveConfig config) {
         objectiveConfigs.add(config);
+        objectiveConfigs.sort(Comparator.comparing((ObjectiveConfig op) -> op.getLevel()));
+
     }
 
     /// <summary>
@@ -158,6 +159,8 @@ public class GlobalParam {
                 return;
             }
         }
+        objectiveConfigs.sort(Comparator.comparing((ObjectiveConfig op) -> op.getLevel()));
+
     }
 
     /// <summary>

src/main/java/com/aps/entity/basic/ObjectiveConfig.java

@@ -18,6 +18,11 @@ public class ObjectiveConfig {
      */
     private boolean enabled;
 
+    /**
+     * 越小越好
+     */
+    private boolean isMinimize;
+
     /**
      * 层级，层级越低优先级越高，1为最高
      */
@@ -28,6 +33,11 @@ public class ObjectiveConfig {
      */
     private double weight;
 
+    /**
+     * 值
+     */
+    private double val;
+
     /**
      * 构造函数
      */

src/main/java/com/aps/service/Algorithm/FitnessCalculator.java

@@ -2,7 +2,10 @@ package com.aps.service.Algorithm;
 
 import com.aps.entity.Algorithm.Chromosome;
 import com.aps.entity.Algorithm.ObjectiveWeights;
+import com.aps.entity.basic.GlobalParam;
+import com.aps.entity.basic.ObjectiveConfig;
 
+import java.util.ArrayList;
 import java.util.List;
 
 /**
@@ -10,6 +13,84 @@ import java.util.List;
  * 时间：2025-11-24
  */
 public class FitnessCalculator {
+
+    /**
+     * 多目标适应度计算（加权求和）
+     */
+    public double[] calculateFitness(Chromosome chromosome, GlobalParam param) {
+
+        ArrayList<Double> list = new ArrayList<>();
+        double fitness=0;
+        int i=0;
+        int level=-1;
+        for (ObjectiveConfig config : param.getObjectiveConfigs()) {
+            if( config.isEnabled()) {
+
+                if(level==-1)
+                {
+                    level=config.getLevel();
+                }else if(level!=config.getLevel())
+                {
+                    list.add(fitness);
+                    fitness=0;
+                    level=config.getLevel();
+                }
+
+
+                double obj = chromosome.getObjectives()[i];
+                // 防御：处理负数（目标值应为非负，若为负则兜底为0）
+                obj = Math.max(0, obj);
+                //转化为0,1 越靠近1 越优
+                double val = 0;
+                if (config.isMinimize()) {
+                    // ============================
+                    // 最小化目标：越小越优 → 归一化后越大越优，越优
+                    // ============================
+                    //0/1+ 0 =0 min max 最优
+                    //1/1+ 1 =0.5 min max 最优
+                    //1/1+ 2 =0.33
+                    //1/1+ 3 =0.25
+                    val = 1.0 / (1.0 + obj);
+                } else {
+                    // ============================
+                    // 最大化目标：越大越优 → 反转成 越大越优
+                    // ============================
+                    //  1/1+ 1 = 0.5
+                    //  2/1+ 2 = 0.66
+                    //  3/1+ 3 =  0.75 max  max 最优
+                    val = obj / (1.0 + obj);
+                }
+                // 兜底：避免极端情况归一化值为0（最小设为1e-6）
+               // val = Math.max(val, 1e-6);
+                if (!param.isPureNSGAIIMode()) {
+                    val = val * config.getWeight();
+                }
+                fitness += val;
+                i++;
+            }
+        }
+        list.add(fitness);
+            // 归一化（假设最大可能值，实际应根据问题规模调整）
+
+            //  makespan = 10 → val = 1/11 = 0.09
+            //  makespan = 20 → val = 1/21 = 0.045
+            //  makespan = 100 → val = 1/101 = 0.009
+            // 值越大 归一化 后越小, 越大越好
+            //原目标（工期、延迟、负载）越小 → 生产越好
+            //原目标越小，归一化后 → f1、f2、f3 越大
+
+//            方案 A：工期短、延迟小 → 很好
+//                makespan=10 → f1=1/(11)=0.09
+//            fitness ≈ 0.08
+//            方案 B：工期长、延迟大 → 很差
+//                makespan=100 → f1=1/(101)=0.009
+//            fitness ≈ 0.01
+
+        // 适应度值（越大越好）
+        return list.stream().mapToDouble(Double::doubleValue).toArray();
+    }
+
+
     /**
      * 多目标适应度计算（加权求和）
      */
@@ -29,11 +110,25 @@ public class FitnessCalculator {
            // 1/1+1 0.5
            // 1/1+2 0.3
            // 1/1+3  0.25
-            // 值越大 归一化 后越小
+
+          //  makespan = 10 → val = 1/11 = 0.09
+          //  makespan = 20 → val = 1/21 = 0.045
+          //  makespan = 100 → val = 1/101 = 0.009
+            // 值越大 归一化 后越小, 越大越好
+            //原目标（工期、延迟、负载）越小 → 生产越好
+            //原目标越小，归一化后 → f1、f2、f3 越大
+
+//            方案 A：工期短、延迟小 → 很好
+//                makespan=10 → f1=1/(11)=0.09
+//            fitness ≈ 0.08
+//            方案 B：工期长、延迟大 → 很差
+//                makespan=100 → f1=1/(101)=0.009
+//            fitness ≈ 0.01
         }
         // 适应度值（越大越好）
         return fitness;
     }
+
     /**
      * 计算种群适应度标准差（用于参数微调）
      * @param population 染色体种群数组

src/main/java/com/aps/service/Algorithm/GeneticDecoder.java

@@ -21,9 +21,7 @@ import java.time.Duration;
 import java.time.LocalDateTime;
 import java.time.temporal.ChronoUnit;
 import java.util.*;
-import java.util.concurrent.CompletableFuture;
-import java.util.concurrent.ConcurrentHashMap;
-import java.util.concurrent.CopyOnWriteArrayList;
+import java.util.concurrent.*;
 import java.util.concurrent.atomic.AtomicInteger;
 import java.util.stream.Collectors;
 import java.util.concurrent.locks.ReentrantLock;
@@ -78,6 +76,36 @@ public class GeneticDecoder {
        // this.orderMaxID = orders.stream().mapToInt(Order::getId).max().orElse(0);
     }
 
+    private final ExecutorService decodeExecutor = new ThreadPoolExecutor(
+            Runtime.getRuntime().availableProcessors() - 1, // 核心线程数=CPU-1，无切换开销
+            Runtime.getRuntime().availableProcessors() - 1, // 最大线程数=核心数
+            0L, TimeUnit.MILLISECONDS,
+            new ArrayBlockingQueue<>(200), // 有界队列，避免内存溢出
+            new ThreadPoolExecutor.CallerRunsPolicy() // 任务满了主线程执行，不丢失任务
+    );
+
+    public   void  Chromosomedecode(List<Chromosome> population)
+    {
+        FileHelper.writeLogFile("解码---------------"+population.size() );
+
+        boolean ismore=false;
+        if(ismore) {
+            CompletableFuture.allOf(population.stream()
+                            .map(chromosome -> CompletableFuture.runAsync(() -> decodeChromosomeWithCache(chromosome), decodeExecutor))
+                            .toArray(CompletableFuture[]::new))
+                    .join();
+
+        } else  {
+            if (population != null && population.size() > 0) {
+                population.forEach(chromosome -> {
+                    decodeChromosomeWithCache( chromosome);
+                });
+            }
+        }
+
+
+    }
+
 //    public Chromosome decodeChromosomeWithCache(Chromosome chromosome) {
 //        String cacheKey = createCacheKey(chromosome);
 //       // FileHelper.writeLogFile("解码-----------开始-------"+chromosome.getID()+":"+cacheKey );
@@ -345,18 +373,18 @@ if(finishedOrder==null||finishedOrder.size()==0)
 //
 //        // 根据工序数量和CPU核心数决定是否使用内部并行
         if (operationCount > 500 && cpuCores > 4) {
-            // 使用并行处理
-            FileHelper.writeLogFile("使用并行处理 _s");
+            // 使用设备并行处理
+          //  FileHelper.writeLogFile("使用并行处理 _s");
             parallelDecodeByMachine(chromosome);
-            FileHelper.writeLogFile("使用并行处理 _e");
+          //  FileHelper.writeLogFile("使用并行处理 _e");
       }
-//        else {
+        else {
             // 使用串行处理
 //            FileHelper.writeLogFile("使用串行处理 _s");
-//            serialDecode(chromosome);
+            serialDecode(chromosome);
 //            FileHelper.writeLogFile("使用串行处理 _e");
 
-        //}
+        }
     }
     /**
      * 染色体解码为调度方案
@@ -555,6 +583,8 @@ if(finishedOrder==null||finishedOrder.size()==0)
         Map<Long, CopyOnWriteArrayList<GAScheduleResult>> machineTasksCache = new HashMap<>();
         Map<Long, Machine> machineIdMap = chromosome.getMachines().stream()
                 .collect(Collectors.toMap(Machine::getId, m -> m));
+        // 构建工序ID索引：opId -> Entry（O(1)查找）
+        Map<Integer, Entry> opIdIndex = buildOpIdIndex(chromosome);
 
         for (GlobalOperationInfo globalOp : globalOpList) {
             int globalOpId = globalOp.getGlobalOpId();
@@ -590,7 +620,8 @@ if(finishedOrder==null||finishedOrder.size()==0)
         FileHelper.writeLogFile("预处理工序信息 _s");
         // 预处理工序信息
         for (int opId : topologicalOrder) {
-            Entry op = findOperationById(chromosome, opId);
+           // Entry op = findOperationById(chromosome, opId);
+            Entry op = opIdIndex.get(opId);
             if (op != null) {
                 // 假设已经为工序分配了设备
 
@@ -610,7 +641,8 @@ if(finishedOrder==null||finishedOrder.size()==0)
         FileHelper.writeLogFile("并行处理 _s");
         for (int opId : topologicalOrder) {
             // 获取当前工序的依赖
-            Entry op = findOperationById(chromosome, opId);
+           // Entry op = findOperationById(chromosome, opId);
+            Entry op = opIdIndex.get(opId);
             List<Integer> dependencies = dependencyGraph.getOrDefault(opId, Collections.emptyList());
 
             // 等待所有依赖完成
@@ -661,10 +693,8 @@ if(finishedOrder==null||finishedOrder.size()==0)
         // 1. 获取工序处理顺序
         List<Integer> operationSequencing = chromosome.getOperationSequencing();
 
-        List<GlobalOperationInfo> globalOpList=chromosome.getGlobalOpList();
-        List<Entry> allOperations=chromosome.getAllOperations();
-
-
+        List<GlobalOperationInfo> globalOpList = chromosome.getGlobalOpList();
+        List<Entry> allOperations = chromosome.getAllOperations();
 
 
         // 2. 构建工序依赖图
@@ -696,7 +726,7 @@ if(finishedOrder==null||finishedOrder.size()==0)
 
             // 获取选择的设备ID和加工时间
             MachineOption selectedMachine = optionalMachines.get(machineSeq - 1);
-            OpMachine opMachine=new OpMachine();
+            OpMachine opMachine = new OpMachine();
             opMachine.setGroupId(groupId);
             opMachine.setSequence(sequence);
             opMachine.setMachineId(selectedMachine.getMachineId());
@@ -850,11 +880,11 @@ if(finishedOrder==null||finishedOrder.size()==0)
             // 工序超过500的设备直接在主线程串行处理
             if (ops.size() > 500) {
                 //   FileHelper.writeLogFile("设备 " + machineId + " 有 " + ops.size() + " 个工序，直接串行处理");
-                processMachineOps(chromosome, machineId, ops, dependencyGraph, opCompleted, lock,opMachineKeyMap,machineIdMap,machineTasksCache);
+                processMachineOps(chromosome, machineId, ops, dependencyGraph, opCompleted, lock, opMachineKeyMap, machineIdMap, machineTasksCache);
             } else {
                 // 工序少的设备创建线程处理
                 Thread thread = new Thread(() -> {
-                    processMachineOps(chromosome, machineId, ops, dependencyGraph, opCompleted, lock,opMachineKeyMap,machineIdMap,machineTasksCache);
+                    processMachineOps(chromosome, machineId, ops, dependencyGraph, opCompleted, lock, opMachineKeyMap, machineIdMap, machineTasksCache);
                 });
                 thread.start();
                 threads.add(thread);
@@ -872,7 +902,7 @@ if(finishedOrder==null||finishedOrder.size()==0)
         // 9. 等待所有设备处理完成
         //  CompletableFuture.allOf(allMachineFutures.toArray(new CompletableFuture[0])).join();
 
-        if(chromosome.getReOrderids()!=null&&chromosome.getReOrderids().size()>0) {
+        if (chromosome.getReOrderids() != null && chromosome.getReOrderids().size() > 0) {
             chromosome.getOperationSequencing().removeIf(t -> chromosome.getReOrderids().contains(t));
         }
 
@@ -1051,8 +1081,19 @@ if(finishedOrder==null||finishedOrder.size()==0)
 
 
 
+
         return result;
     }
+    /**
+    * 构建工序ID索引：opId -> Entry（O(1)查找）
+   */
+    private Map<Integer, Entry> buildOpIdIndex(Chromosome chromosome) {
+        Map<Integer, Entry> index = new HashMap<>();
+        for (Entry op : chromosome.getAllOperations()) {
+            index.put(op.getId(), op);
+        }
+        return index;
+    }
     /**
      * 根据工序ID查找工序
      */
@@ -2444,6 +2485,10 @@ if(MaterialRequirements==null||MaterialRequirements.size()==0)
         }
 
         chromosome.setObjectives(Objectives);
+        FitnessCalculator fitnessCalculator=new FitnessCalculator();
+        if (chromosome.getFitnessLevel()==null) {
+            chromosome.setFitnessLevel(fitnessCalculator.calculateFitness(chromosome, _globalParam));
+        }
     }
 
     private double calculateTotalFlowTime(Chromosome chromosome) {

src/main/java/com/aps/service/Algorithm/ParallelLocalSearch.java

@@ -48,10 +48,10 @@ public class ParallelLocalSearch {
         });
 
         // 添加模拟退火任务
-        tasks.add(() -> {
-            SimulatedAnnealing sa = new SimulatedAnnealing(globalParam, allOperations, globalOpList, fitnessCalculator, objectiveWeights);
-            return sa.search(chromosome, decoder, param);
-        });
+//        tasks.add(() -> {
+//            SimulatedAnnealing sa = new SimulatedAnnealing(globalParam, allOperations, globalOpList, fitnessCalculator, objectiveWeights);
+//            return sa.search(chromosome, decoder, param);
+//        });
 
         // 执行所有任务
         List<Chromosome> results = new ArrayList<>();

src/main/java/com/aps/service/plan/PlanResultService.java

@@ -210,7 +210,7 @@ public class PlanResultService {
                     .collect(Collectors.toList());
             globalParam.setPureNSGAIIMode(false);
             // 5. 执行调度算法
-            GeneticAlgorithm scheduler =new GeneticAlgorithm(globalParam,machines,orders,Materials1,materialIds,machineScheduler,entryRel,materialRequirementService,_sceneService,SceneId); //new GeneticAlgorithm(products, machines, orders, machineScheduler);
+            HybridAlgorithm scheduler =new HybridAlgorithm(globalParam,machines,orders,Materials1,materialIds,machineScheduler,entryRel,materialRequirementService,_sceneService,SceneId); //new GeneticAlgorithm(products, machines, orders, machineScheduler);
             param.initAdaptiveParams(entrys.size());
             double[] customWeights = new double[] { 0.4, 0.1, 0.1, 0.1, 0.3 }; // 延迟时间权重提升到0.5
             //完工时间、总流程时间、总换型时间、机器负载标准差、延迟时间

src/test/java/com/aps/demo/PlanResultServiceTest.java

@@ -39,7 +39,9 @@ public class PlanResultServiceTest {
       //  TestSortService  sortService=new TestSortService();
      //  sortService.test1();
        // nsgaiiUtils.Test();
-      planResultService.execute2("0428340BB4F540938F1FB5599F03E8A4");
+    //  planResultService.execute2("0428340BB4F540938F1FB5599F03E8A4");//2000
+        planResultService.execute2("C8B533BD8944405B9A2F8823C575C204");//500
+      //  planResultService.execute2("EFDD34E4B5BC434BAEAE6A84DFCD4E7B");//20
       //  planResultService.execute2("00E0C5D3E4AD4F36B56C39395906618D");
 
       //   planResultService.execute2("92BB773E1E2447C99D8176C991D5C9D2");