我有一个相当小的 MRP 应用程序，它由销售订单簿、材料清单和材料资源计划报告组成。
销售订单簿只是项目、数量和到期日的列表。物料清单是每个项目、该项目的子组件和数量的平面列表，如下所示：[ParentItemID] [ComponentItemID] [Qty]
然后我还添加了我的采购订单，以便用户可以看到运行余额。
它（应该）计算起来相当简单。
MRP 报告在功能上运行良好，但是运行大约需要 90 秒，这显然让我的用户感到沮丧。我目前正在使用队列来存储数据，我的整个代码如下所示：
public void MRPCalcBtn_Click(object sender, EventArgs e)
//Calculate MRP Report
{
//kick off a timer
DateTime startTime = DateTime.Now;
//Flip the panels to hide the button
panel1.Visible = false;
panel2.Visible = true;
Page.Title = "MRP Report"; //set the title so we can tell if it's finished from the task bar
//Create DataTable to hold items
var dt = new DataTable();
dt.Columns.Add("ItemID", System.Type.GetType("System.String"));
dt.Columns.Add("QtyIn", System.Type.GetType("System.Decimal"));
dt.Columns.Add("QtyOut", System.Type.GetType("System.Decimal"));
dt.Columns.Add("DueDate", System.Type.GetType("System.DateTime"));
var db = new CS3Entities(); // should these be calls to DAL?
//ADD SALES ORDERS
//---------------------------------------------------------------------------------------------
//read orderbook in
var orderBook = (from ob in db.SalesOrderItems
where ob.QtyOutstanding > 0
//where ob.ItemID == "9783B" //whilst testing, remove in production
select new { ob.ItemID, ob.QtyOutstanding, ob.DueDate });
//create a queue to store temporary work, the line items are Orders
var q1 = new Queue<OrderItem>();
foreach (var item in orderBook) //loop through the order book
{
//Copy order book to queue to start off with
var orderBookArray = new OrderItem
{
ItemID = item.ItemID,
QtyIn = 0,
QtyOut = item.QtyOutstanding,
DueDate = item.DueDate
};
q1.Enqueue(orderBookArray); //adds line item to queue
}
//start calculating the BoM items
try
{
do //main working loop
{
var testItem = q1.Peek(); //pull out next line item in queue
var testItemID = testItem.ItemID; //pull out ItemID from line item
var testQty = testItem.QtyOut; //pull out Qty from line item
var testDueDate = testItem.DueDate; //pull out ItemID from line item
//determines Bill OF Materials for each item
var bills = (
from b in db.BillOfMaterials
where b.ItemID == testItemID
select new { b.ComponentItemID, b.Qty }
);
if (bills.Any()) //if line item has any sub components
{
//add each sub comopnent to the queue
foreach (var bill in bills)
{
var bomArray = new OrderItem
{
ItemID = bill.ComponentItemID,
QtyOut = bill.Qty*testQty,
QtyIn = 0,
DueDate = testDueDate
};
q1.Enqueue(bomArray);
}
q1.Dequeue(); //Remove item from queue after all sub components have been added
}
else //if no items in Bill Of Materials, must be "leaf" component, add it to "final" table
{
AddToItems(dt, testItemID, 0, testQty, testDueDate); //add item to data table as QtyOut
q1.Dequeue(); //Remove item from queue
}
} while (q1.Count != 0); //end of do loop, until queue is empty
}
catch
{
ErrorLbl.Text = "ERROR"; //if something goes wrong
}
//Queue should now be empty and DataTable dt holds all final items for purchase
//ADD PURCHASE ORDERS
//---------------------------------------------------------------------------------------------
//read purchase orders in
var purchaseOrderItems = (from poi in db.PurchaseOrderItems
join po in db.PurchaseOrders on poi.PurchaseOrderID equals po.PurchaseOrderID
where po.PurchaseOrderCancelled != true &&
po.PurchaseOrderDeleted != true &&
poi.Balance > 0 &&
poi.LineItemComplete != true
orderby poi.DueDate ascending
select new { poi.ItemID, poi.Balance, poi.DueDate }).ToList(); //Is a list faster than the database calls?
foreach (var item in purchaseOrderItems) //loop through the purchase orders and append them to the temp table
{
//add purchase order items to data table as QtyIn
AddToItems(dt, item.ItemID, Convert.ToDecimal(item.Balance), 0, Convert.ToDateTime(item.DueDate));
}
//SORT RESULTS
//---------------------------------------------------------------------------------------------
//Sort and group the raw data
var sorteditems = //sort and group the data table to show sum of qty per day per item
(from p in dt.AsEnumerable()
orderby p.Field<string>("ItemID") , p.Field<DateTime>("DueDate")
group p by new {ItemID = p.Field<string>("ItemID"), DueDate = p.Field<DateTime>("DueDate")}
into g
select new
{
g.Key.ItemID,
g.Key.DueDate,
QtyIn = g.Sum(p => p.Field<decimal>("QtyIn")),
QtyOut = g.Sum(p => p.Field<decimal>("QtyOut"))
}).ToList();
//Create second DataTable to hold sorted and grouped report items
//tried to use same datatable dt but it didn't work, no hardship to instantiate another
var dt2 = new DataTable();
dt2.Columns.Add("ItemID", Type.GetType("System.String"));
dt2.Columns.Add("QtyIn", Type.GetType("System.Decimal"));
dt2.Columns.Add("QtyOut", Type.GetType("System.Decimal"));
dt2.Columns.Add("DueDate", Type.GetType("System.DateTime"));
foreach (var item in sorteditems) //loop through the sorted results
{
//Move the sorted and grouped items into the datatable
AddToItems(dt2, item.ItemID, Convert.ToDecimal(item.QtyIn), Convert.ToDecimal(item.QtyOut), Convert.ToDateTime(item.DueDate));
}
Session["MRPTable"] = dt2; //save datatable to session for recall later by repeater
//Set repeater datasource to unique item list with details
var itemslist = from si in sorteditems
group si by new { si.ItemID } into sig //equivilent to DISTINCT
join i in db.Items
on sig.Key.ItemID equals i.ItemID
select new
{
i.ItemID,
i.ItemName,
i.StocktakeCost,
i.ItemROP,
i.ItemROQ,
i.ItemUOM,
i.ItemStock
};
Repeater1.DataSource = itemslist; //note repeater contains a GridView that contains the line items
Repeater1.DataBind(); //Fire the repeater
//time finished
DateTime finishTime = DateTime.Now;
//Set the start time label
TimeStart.Text = startTime.TimeOfDay.ToString();
//Set finish time
TimeFinish.Text = finishTime.TimeOfDay.ToString();
TimeTaken.Text = (finishTime - startTime).Minutes.ToString(CultureInfo.InvariantCulture) + " Minute(s), " + (finishTime - startTime).Seconds.ToString(CultureInfo.InvariantCulture) + " Seconds";
//Add to activity log
ActivityDAL.AddNewLog("MRP Report", "Generated", TimeTaken.Text, "N/A", HttpContext.Current.User.Identity.Name);
}
谁能看到我在哪里可以提高报告的速度以及我在哪里犯了任何明显的错误。
正如我所说，报告工作正常，但速度很慢。
任何帮助表示赞赏