Production Scheduling for Construction Chemicals

Production scheduling for construction chemicals means sequencing dry powder blending and liquid mixing campaigns across shared packaging lines while managing colour-driven and chemistry-crossover changeovers. Schantt models the hybrid batch-and-flow pipeline — batch mixers and reactors feeding parallel flow filling stations — with directional changeovers, per-class routings that skip stages they do not need, and shift-aware calendars that expand for peak construction season. This guide walks through configuring a two-product-class, four-stage scenario so you can apply the same modeling patterns to your own plant.

This guide follows a fictional composite company built from industry research on construction chemicals; all names, parameters, and figures are illustrative.

Industry context

Construction chemicals manufacturing follows a split-flowshop topology: dry powder products — cementitious mortars, tile adhesives, grouts, and joint compounds — are blended in batch ribbon mixers and conveyed to packaging, while liquid products — concrete admixtures, waterproofing membranes, and surface treatments — are reacted in stirred jacketed reactors, homogenised through high-shear dispersers, and then filled into drums, pails, or IBC totes. Both streams converge on a shared packaging hall with parallel filling lines serving different formats. The industry serves building-material distributors, ready-mix concrete plants, and specialised contractor networks with products that must meet compressive-strength, viscosity, and regulatory standards.

A typical mid-market facility operates 5–12 machines across 3–5 production stages, managing 100–500 SKUs across 3–5 product classes. Colour variants within a product class — white, light grey, dark grey — require directional changeovers: light-to-dark transitions take 15–30 minutes, while dark-to-light reversals can take 30–45 minutes of filler-head cleaning. Chemistry-crossover changeovers between incompatible product families, such as cementitious to admixture, add 30–45 minutes per transition for flushing and purging the shared line. Production follows a mixed make-to-stock and make-to-order model, roughly 55 % make-to-stock and 45 % make-to-order, with seasonal demand surging 30–50 % in spring and summer as construction activity peaks. A two-level changeover hierarchy applies on shared packaging lines — chemistry-level transitions between product families subsume colour-level transitions within a family.

The dominant scheduling challenge is sequencing jobs to minimise total changeover time across shared mixers and filling lines while handling rush order insertions and seasonal capacity swings. Planners typically manage 10–15 jobs per day with spreadsheets and whiteboard Gantts — a manual process that makes changeover optimisation and line-balancing trade-offs difficult to quantify. Weekly capacity almost doubles between off-peak periods (approximately 40 hours per week) and peak construction season (approximately 78 hours per week on double shifts), adding another layer of complexity to the scheduling task.

Crestline Materials runs 90 people at a 5,200 m² facility, making 2 product classes across 4 production stages, scheduled by a 3-person planning team.

Process overview

flowchart LR
    DB["Dry Blending<br/>(BATCH)"] --> PF["Packaging & Filling<br/>(FLOW)"]
    LM["Liquid Mixing & Reaction<br/>(BATCH)"] --> HG["Homogenisation<br/>(BATCH)"]
    HG --> PF

Four production stages across two product-class routings converge on a shared packaging hall.

Skip-routing note: Cementitious mortars route through Dry Blending directly to Packaging & Filling, skipping Liquid Mixing and Homogenisation. Concrete admixtures start at Liquid Mixing and route through Homogenisation to Packaging, skipping Dry Blending entirely. Transfer times bridge the gaps — a 10-minute surge hopper gravity feed from Dry Blending to Packaging, a 20-minute intermediate holding tank transfer from Liquid Mixing to Homogenisation, and a 15-minute day tank buffer from Homogenisation to Packaging. The admixture routing also uses partial transfer on the Liquid Mixing to Homogenisation leg, enabling the homogeniser to begin work on the first 3,000 kg while the reactor continues processing the remaining half-batch.

Scheduling challenges and how Schantt handles them

Crestline's schedule is driven by make-to-stock demand for standard mortars and admixtures — roughly 55 % of volume — supplemented by make-to-order rush orders that arrive with tight delivery windows. The remaining 45 % of volume consists of custom-colour mortar runs and performance-grade admixture orders that the planning team must insert into an already-packed schedule. If your plant runs a different demand profile — pure make-to-order or pure make-to-stock — the same modeling patterns apply, though the mode choice may shift. The optimizer minimises total production time and schedules forward from a start date; this guide assumes a one-to-two-week planning horizon during peak construction season. Auto mode sequences make-to-stock jobs to minimise changeover time; Semi-Auto mode lets the planner insert make-to-order rush orders at a chosen position with earliest-start constraints, preserving the planner's sequencing judgement while quantifying the time impact.

What Schantt handles well

• Sequential multi-stage production — Construction chemicals follow ordered routes through dry blending or liquid mixing then packaging. Schantt models the ordered stage sequence, per-class routing, and forward-only transfer times so each job flows step-by-step.

• Multi-machine stages (parallel machines) — Packaging departments operate parallel filling lines for bags, pails or drums, and IBC totes. Schantt groups machines under a single stage, so the algorithm assigns jobs to the best available line in Auto and Semi-Auto modes.

• Mixed batch-and-flow pipelines — Dry blending and liquid mixing are batch stages; packaging is a flow stage. Schantt models both in one route, correctly timing the batch-to-flow handoff and surfacing wait-material pauses when packaging outruns upstream supply.

• Multi-product routing with stage skipping — Product classes skip stages they do not need. Mortars skip liquid stages, while admixtures skip dry blending. Schantt's per-class routing includes only the stages each class requires, and bridging transfer times connect the gap.

• Sequence-dependent changeovers — Colour-driven and chemistry-crossover changeovers are directional. Schantt models directional changeover times per machine per product-class pair, so the optimizer favours sequences that minimise total setup time.

• Shift-aware availability and calendar exceptions — Plants run single- or double-shift patterns that expand during peak season and contract in winter. Schantt models working calendars with shift patterns, holiday exceptions, and downtime windows for planned maintenance.

How Schantt handles each challenge

1. Chemistry-crossover changeovers between product families.
- Crestline's pail-or-drum line is the only machine shared by both product classes. Every transition between a mortar campaign and an admixture campaign on this line incurs a 30-minute cleaning and purge cycle for mortar to admixture, or a 45-minute cycle for admixture to mortar. With 15–25 job switches per week across the shared line, total changeover time accounts for 8–15 % of available runtime — the equivalent of losing half a shift or more every week to filler-head cleaning and flushing.
- Schantt models these as directional changeover times on the shared line — one entry per direction with the exact durations from the dataset. When the optimizer sequences jobs in Auto mode, it naturally favours patterns that group same-class jobs together to minimise the number of class-crossing transitions. The planner can see the total changeover time per sequence and compare alternatives in Semi-Auto mode, enabling an evidence-based decision about whether a rush insertion is worth the cleaning time it will trigger.

2. Batch-to-flow handoff mismatch between mixing and packaging.
- Dry blending produces discrete 2,000 kg batches every 10 minutes. The bag line consumes at 10,000 kg per hour, which means it can exhaust a full blender batch in roughly 12 minutes — barely longer than the blend cycle itself. When the surge hopper empties between batches, or when the reactor-to-homogeniser-to-IBC cascade starves, the filling line idles 20–40 minutes. This starvation pattern occurs two to three times per week in peak season, adding up to 1–2 hours of unplanned downtime.
- Schantt models the batch-to-flow pipeline natively: the algorithm knows batch stages produce discrete loads at fixed intervals while flow stages consume continuously. When a downstream flow stage has consumed all available upstream material and the next batch is not yet complete, the schedule automatically inserts a wait-material pause — making starvation visible and quantifiable on the timeline. The planner can adjust batch overlap, stagger shift start times, or increase buffer transfer quantities to reduce the gap between supply and consumption.

3. Parallel packaging line balancing under mixed make-to-stock and make-to-order load.
- Three packaging lines serve two product classes with different format preferences. The bag line runs near capacity, at approximately 95 % utilisation, on make-to-stock mortars; the IBC station runs admixture make-to-order orders at 60–80 %. The pail-or-drum line serves as the swing resource, but each class switch adds 30–45 minutes of cleaning time. The planning team's heuristic — bag line for mortars, IBC for admixtures, pail-or-drum for overflow — breaks down when a large admixture order exceeds IBC capacity.
- Schantt's multi-machine stage groups all three packaging lines under one stage. In Auto mode, the algorithm considers each line's throughput, current utilisation, and the changeover time of assigning a job to the shared line versus a dedicated line. It quantifies the trade-off between splitting an order across lines versus accepting a changeover — a calculation that spreadsheets cannot support.

4. Make-to-stock and make-to-order sequencing tension and colour reversals in peak season.
- Make-to-stock standard colours — white and light grey — sequence with minimal cleaning time between them. Make-to-order rush orders — custom colours, dark shades — arrive with delivery pressure that forces suboptimal insertions. A single dark-to-light reversal on the shared packaging line adds 30–45 minutes of cleaning. During peak season, planners handle 6–10 such rush insertions per week, manually re-sequencing the board — 1–2 hours of the head planner's daily work.
- Schantt's directional changeover times encode the colour penalty: light-to-dark transitions take 15–30 minutes, dark-to-light take 30–45 minutes. In Auto mode, the optimizer penalises reversals and naturally produces light-to-dark sequences. For rush make-to-order orders, Semi-Auto mode lets the planner insert the job at a chosen position and immediately see the added changeover time on the schedule, making the trade-off explicit.

What to model in Schantt

A Schantt model for this scenario uses five first-class entities at the counts below.

Step-by-step setup

1. Create the stages in order. Add four stages in sequence: Dry Blending as a batch stage, Liquid Mixing & Reaction as a batch stage, Homogenisation as a batch stage, and Packaging & Filling as a flow stage. On each stage's detail page, set the transfer times to the next stage that the routing uses:

• Liquid Mixing to Homogenisation: 20 minutes
• Homogenisation to Packaging: 15 minutes
• Bridging transfer from Dry Blending to Packaging: 10 minutes (for the mortar class that skips the intermediate stages)

2. Add the machines to each stage. Assign two horizontal ribbon blenders to Dry Blending, two stirred jacketed reactors to Liquid Mixing & Reaction, one high-shear rotor-stator disperser to Homogenisation, and three stations to Packaging & Filling — a bag line, a combined pail-and-drum line, and an IBC fill station.

3. Create the product classes and define each class's routing. Add Cementitious Mortars and Concrete Admixtures as the two product classes. For mortars, the routing includes only Dry Blending then Packaging (stages 1 and 4) — set it to skip Liquid Mixing and Homogenisation entirely. For admixtures, the routing includes Liquid Mixing, Homogenisation, then Packaging (stages 2, 3, and 4) — set it to skip Dry Blending. These skip-routing definitions are what enable the algorithm to model each product class through only the stages it actually needs, avoiding wasteful idle time on irrelevant equipment. Enable partial transfer on the admixture's Liquid Mixing to Homogenisation leg with a minimum charge of 3,000 kg, reflecting that the reactor batch can begin feeding the disperser before the full 6,000 kg is complete.

4. Add the products. Create one representative product per class: Standard Tile Adhesive for cementitious mortars and PCE Superplasticizer 30 % for concrete admixtures.

5. Set each machine's capacity parameters and changeovers. For batch machines, enter the cycle duration and batch capacity. Each dry blender runs at 2,000 kg per batch with a 10-minute cycle. Each reactor runs at 6,000 kg per batch with a 60-minute cycle. The high-shear disperser runs at 3,000 kg per batch with a 25-minute cycle — note that a full reactor batch at 6,000 kg is split into two disperser charges.

For flow machines, enter the throughput per hour. The bag line handles 440 bags per hour at 25 kg per bag, giving a consumption rate of roughly 11 tonnes per hour. The pail-or-drum line runs at 120 pails per hour for mortar or 40 drums per hour for admixture. The IBC station fills 10 totes per hour at 1,000 L per tote.

On the pail-or-drum line — the only machine shared by both classes — add directional changeovers: 30 minutes for mortar to admixture, 45 minutes for admixture to mortar. All other machines serve a single class and need no changeover entries. Machine eligibility is expressed implicitly: processing-time entries exist only for compatible class-machine pairs, so the dry blenders have entries only for mortars, the reactors and disperser only for admixtures, the bag line only for mortars, and the IBC station only for admixtures.

6. Configure calendars, exceptions, and downtimes. Set the standard single-shift calendar as default — Monday through Thursday 07:00–16:30, Friday 07:00–12:00, weekend non-working — providing approximately 40 hours of weekly capacity. Add a peak-season calendar with expanded double shifts — Monday through Thursday 06:00–22:00, Friday 06:00–16:00, Saturday 06:00–12:00 — providing approximately 78 hours of weekly capacity. Apply the peak-season calendar via a schedule calendar period covering the full Q2–Q3 construction months; the planner controls when to switch, so the transition is a single setting rather than a day-by-day override. Add calendar exceptions for New Year's Day (January 1) and International Workers' Day (May 1). Schedule downtimes for the annual mixer relining on RB-1 (February 15, 07:00–16:30), the quarterly filler head overhaul on PL-1 (March 20, 08:00–15:00), and the year-end plant shutdown (December 24, 12:00 to January 1, 08:00).

For step-by-step instructions on configuring each of these in Schantt, see the Schantt documentation.

Common mistakes

1. A single blanket changeover instead of directional per-pair times. Using one cleaning duration for all transitions on the shared packaging line ignores the asymmetry — a 30-minute transition one way and 45 minutes the other. Fix: Create one directional entry per class pair — mortar to admixture at 30 minutes, admixture to mortar at 45 minutes — so the optimizer correctly accounts for the longer direction.

2. One product class covering divergent routings. Defining mortars and admixtures as a single construction chemicals class forces both through the same stages, eliminating the skip-routing benefit. Fix: Create separate classes for each distinct routing pattern. Only classes with different stage sets should be split, not every colour or grade variant.

3. A stage's machine count that does not match the floor. Listing one packaging machine per class underestimates the parallel-line assignment complexity and prevents the algorithm from balancing load across the bag line, pail-or-drum line, and IBC station. Fix: Register each physical machine as its own entry under the Packaging & Filling stage — three machines with their real throughputs and changeover matrices — so the algorithm can consider all available capacity and choose the best line for each job.

4. Ignoring batch-to-flow physics at the handoff. Setting packaging as a batch stage with a fast cycle time instead of a flow stage with throughput loses the starvation-awareness behaviour. Fix: Model filling stations as flow stages. The natural mismatch between batch supply and continuous consumption will surface wait-material pauses that reveal genuine bottlenecks.

5. Not using schedule calendar periods for seasonal shifts. Hard-coding one shift pattern year-round forces the planner to manually override every day during peak season. Fix: Keep two calendars and apply the peak-season calendar to a date range via a schedule calendar period — the switch is a single setting, not a day-by-day edit.

What a good schedule looks like

A well-configured Schantt schedule replaces manual whiteboard sequencing with optimised production timelines that surface trade-offs explicitly.

Before (spreadsheet and whiteboard): Crestline's planning team manually sequences jobs using colour-group heuristics — lightest to darkest, mortar campaigns before admixture runs — but rush orders cascade into changeover reversals that the team cannot quantify until the cleaning time is already spent. Starvation at the packaging hall goes undetected until a filling line stops and an operator calls the planner. The head planner spends 1–2 hours per day re-sequencing the whiteboard for rush insertions and cannot compare alternative sequences without completely re-drawing the board, which is rarely practical during a live shift.

After (Schantt Auto and Semi-Auto modes): In Auto mode, the optimizer sequences make-to-stock jobs to minimise total changeover time — grouping same-class campaigns, favouring light-to-dark colour progression, and reducing reversal penalties automatically. Batch-to-flow starvation is visible as wait-material pauses on the timeline, letting the planner adjust batch overlap or stagger starts. For make-to-order rush orders, Semi-Auto mode inserts the job at the planner's chosen position and immediately shows the added changeover time in the total — making the impact of each insertion explicit. The seasonal calendar switch between off-peak and peak shift patterns is a single setting per date range, not a daily manual override. The planner gains back the 1–2 hours spent on manual re-sequencing and can evaluate sequence alternatives in minutes.