SQL one to many relation table design, the right way - sql

Question

Plot: I need to book an Order which relies on 3 different type of Factory.
I have individual tables for both order and Factory. Now I need to make a one to many relations between Order and Booking Factory.
Order Table:
CREATE TABLE [dbo].[tbl_OrderInformation](
[OrderInformationId] [int] IDENTITY(1,1) NOT NULL,
[OrderId] [int] NOT NULL,
[OrderNo] [nvarchar](50) NOT NULL,
CONSTRAINT [PK_tbl_OrderInformation] PRIMARY KEY CLUSTERED
( [OrderInformationId] ASC)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE =
OFF, IGNORE_DUP_KEY = OFF,
ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]
Factory Table:
CREATE TABLE [dbo].[tbl_Factory](
[FactoryId] [int] IDENTITY(1,1) NOT NULL,
[FactoryName] [nvarchar](50) NOT NULL,
[FactoryType] [nvarchar](50) NOT NULL,
CONSTRAINT [PK_tbl_Factory] PRIMARY KEY CLUSTERED
( [FactoryId] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF,
ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]
Sample Order Data
Sample Factory Data
Now, an Order relies on multiple garments, dyeing, and printing Factory.
Suppose, Order C101 relies on Garments-A, Dyeing-A, Printing-A, Printing-B, Printing-C.
Now, I can design OrderBooking table in 2 ways.
CREATE TABLE [dbo].[tbl_OrderBooking_1](
[OrderBookingId] [INT] IDENTITY(1,1) NOT NULL,
[OrderId] [INT] NOT NULL,
[FactoryId] [INT] NULL,
[FactoryType] [NVARCHAR](50) NULL,
CONSTRAINT [PK_tbl_OrderBooking_1] PRIMARY KEY CLUSTERED
(
[OrderBookingId] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF,
ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]
This the data will look like below:
And The second way,
CREATE TABLE [dbo].[tbl_OrderBooking_2](
[OrderBookingId] [INT] IDENTITY(1,1) NOT NULL,
[OrderId] [INT] NULL,
[garmentsFactoryId] [INT] NULL,
[dyeingFactoryId] [INT] NULL,
[printingFactoryId] [INT] NULL,
CONSTRAINT [PK_tbl_OrderBooking_2] PRIMARY KEY CLUSTERED
(
[OrderBookingId] ASC
)WITH (PAD_INDEX = OFF, STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF,
ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON) ON [PRIMARY]
) ON [PRIMARY]
Here the data will look like,
Now, which approach of designing the OrderBooking table is more accurate and why ?
Please keep in mind that the type of factory is fixed to 3, and OrderBooking table will grow quite large over time thus tend to have heavy read and write operations.

Answer 1

The link table between Order & factory would be the best approach.
You will get better performance and you can create indexes on the numeric columns.
Going forward if you have any new factory then it is also each to insert without any issue.
The link table will help you to align with the Normalization rule as well. so my suggestion is go with that.